The field of neural rendering has seen remarkable progress, driven by advancements in generative models and differentiable rendering techniques. While 2D diffusion has achieved notable success, the development of a unified 3D diffusion pipeline remains an open challenge. This paper presents a novel framework, LN3Diff++, designed to bridge this gap and facilitate fast, high-quality, and versatile conditional 3D generation. Our method leverages a 3D-aware architecture and a variational autoencoder (VAE) to encode input image(s) into a structured, compact 3D latent space. The latent representation is then decoded by a transformer-based decoder into a high-capacity 3D neural field. By training a diffusion model on this 3D-aware latent space, our method achieves superior performance for category-specific 3D generation on ShapeNet and FFHQ, as well as category-free image/text-conditioned 3D generation over Objaverse. Moreover, it surpasses existing 3D diffusion methods in inference speed, requiring no per-instance optimization.

Expressive human pose and shape estimation (EHPS) unifies body, hands, and face motion capture with numerous applications. Despite encouraging progress, current state-of-the-art methods focus on training innovative architectural designs on confined datasets. In this work, we investigate the impact of scaling up EHPS towards a family of generalist foundation models. 1) For data scaling, we perform a systematic investigation on 40 EHPS datasets, encompassing a wide range of scenarios that a model trained on any single dataset cannot handle. More importantly, capitalizing on insights obtained from the extensive benchmarking process, we optimize our training scheme and select datasets that lead to a significant leap in EHPS capabilities. Ultimately, we achieve diminishing returns at 10 M training instances from diverse data sources. 2) For model scaling, we take advantage of vision transformers (up to ViT-Huge as the backbone) to study the scaling law of model sizes in EHPS. To exclude the influence of algorithmic design, we base our experiments on two minimalist architectures: SMPLer-X, which consists of an intermediate step for hand and face localization, and SMPLest-X, an even simpler version that reduces the network to its bare essentials and highlights significant advances in the capture of articulated hands. With Big Data and the large model, the foundation models exhibit strong performance across diverse test benchmarks and excellent transferability to even unseen environments. Moreover, our finetuning strategy turns the generalist into specialist models, allowing them to achieve further performance boosts. Notably, our foundation models consistently deliver state-of-the-art results on seven benchmarks such as AGORA, UBody, EgoBody, and our proposed SynHand dataset for comprehensive hand evaluation.

In-context segmentation has drawn increasing attention with the advent of vision foundation models. Its goal is to segment objects using given reference images. Most existing approaches adopt metric learning or masked image modeling to build the correlation between visual prompts and input image queries. This work approaches the problem from a fresh perspective - unlocking the capability of the latent diffusion model (LDM) for in-context segmentation and investigating different design choices. Specifically, we examine the problem from three angles: instruction extraction, output alignment, and meta-architectures. We design a two-stage masking strategy to prevent interfering information from leaking into the instructions. In addition, we propose an augmented pseudo-masking target to ensure the model predicts without forgetting the original images. Moreover, we build a new and fair in-context segmentation benchmark that covers both image and video datasets. Experiments validate the effectiveness of our approach, demonstrating comparable or even stronger results than previous specialist or visual foundation models. We hope our work inspires others to rethink the unification of segmentation and generation.

Auxiliary-free human video matting methods, which rely solely on input frames, often struggle with complex or ambiguous backgrounds. To address this, we propose MatAnyone, a robust framework tailored for target-assigned video matting. Specifically, building on a memory-based paradigm, we introduce a consistent memory propagation module via region-adaptive memory fusion, which adaptively integrates memory from the previous frame. This ensures semantic stability in core regions while preserving fine-grained details along object boundaries. For robust training, we present a larger, high-quality, and diverse dataset for video matting. Additionally, we incorporate a novel training strategy that efficiently leverages large-scale segmentation data, boosting matting stability. With this new network design, dataset, and training strategy, MatAnyone delivers robust and accurate video matting results in diverse real-world scenarios, outperforming existing methods.

Multiview diffusion models have shown considerable success in image-to-3D generation for general objects. However, when applied to human data, existing methods have yet to deliver promising results, largely due to the challenges of scaling multiview attention to higher resolutions. In this paper, we explore human multiview diffusion models at the megapixel level and introduce a solution called mesh attention to enable training at 10242resolution. Using a clothed human mesh as a central coarse geometric representation, the proposed mesh attention leverages rasterization and projection to establish direct cross-view coordinate correspondences. This approach significantly reduces the complexity of multiview attention while maintaining cross-view consistency. Building on this foundation, we devise a mesh attention block and combine it with keypoint conditioning to create our human-specific multiview diffusion model, MEAT. In addition, we present valuable insights into applying multiview human motion videos for diffusion training, addressing the longstanding issue of data scarcity. Extensive experiments show that MEAT effectively generates dense, consistent multiview human images at the megapixel level, outperforming existing multiview diffusion methods. Code is available at https://johann.wang/MEAT/.

Despite advances in neural rendering, due to the scarcity of high-quality 3D datasets and the inherent limitations of multi-view diffusion models, view synthesis and 3D model generation are restricted to low resolutions with suboptimal multi-view consistency. In this study, we present a novel 3D enhancement pipeline, dubbed 3DENHANCER, which employs a multi-view latent diffusion model to enhance coarse 3D inputs while preserving multi-view consistency. Our method includes a pose-aware encoder and a diffusion-based denoiser to refine low-quality multi-view images, along with data augmentation and a multi-view attention module with epipolar aggregation to maintain consistent, high-quality 3D outputs across views. Unlike existing video-based approaches, our model supports seamless multi-view enhancement with improved coherence across diverse viewing angles. Extensive evaluations show that 3DENHANCER significantly outperforms existing methods, boosting both multi-view enhancement and per-instance 3D optimization tasks.

Recent advances in 3D Gaussian Splatting (3D-GS) have shown remarkable success in representing 3D scenes and generating high-quality, novel views in real-time. However, 3D-GS and its variants assume that input images are captured based on pinhole imaging and are fully in focus. This assumption limits their applicability, as real-world images often feature shallow depth-of-field (DoF). In this paper, we introduce DoF-Gaussian, a controllable depth-of-field method for 3D-GS. We develop a lens-based imaging model based on geometric optics principles to control DoF effects. To ensure accurate scene geometry, we incorporate depth priors adjusted per scene, and we apply defocus-to-focus adaptation to minimize the gap in the circle of confusion. We also introduce a synthetic dataset to assess refocusing capabilities and the model’s ability to learn precise lens parameters. Our framework is customizable and supports various interactive applications. Extensive experiments confirm the effectiveness of our method. Our project is available at https://dof-gaussian.github.io/.

Video restoration poses non-trivial challenges in maintaining fidelity while recovering temporally consistent details from unknown degradations in the wild. Despite recent advances in diffusion-based restoration, these methods often face limitations in generation capability and sampling efficiency. In this work, we present SeedVR, a diffusion transformer designed to handle real-world video restoration with arbitrary length and resolution. The core design of SeedVR lies in the shifted window attention that facilitates effective restoration on long video sequences. SeedVR further supports variable-sized windows near the boundary of both spatial and temporal dimensions, overcoming the resolution constraints of traditional window attention. Equipped with contemporary practices, including causal video autoencoder, mixed image and video training, and progressive training, SeedVR achieves highly-competitive performance on both synthetic and real-world benchmarks, as well as AI-generated videos. Extensive experiments demonstrate SeedVR’s superiority over existing methods for generic video restoration.

Endowing Large Multimodal Models (LMMs) with visual grounding capability can significantly enhance AIs’ understanding of the visual world and their interaction with humans. However, existing methods typically fine-tune the parameters of LMMs to learn additional segmentation tokens and overfit grounding and segmentation datasets. Such a design would inevitably cause a catastrophic diminution in the indispensable conversational capability of general AI assistants. In this paper, we comprehensively evaluate state-of-the-art grounding LMMs across a suite of multimodal question-answering benchmarks, observing drastic performance drops that indicate vanishing general knowledge comprehension and weakened instruction following ability. To address this issue, we present F-LMM—grounding frozen off-the-shelf LMMs in human-AI conversations—a straightforward yet effective design based on the fact that word-pixel correspondences conducive to visual grounding inherently exist in the attention mechanism of well-trained LMMs. Using only a few trainable CNN layers, we can translate word-pixel attention weights to mask logits, which a SAM-based mask refiner can further optimise. Our F-LMM neither learns special segmentation tokens nor utilises high-quality grounded instruction-tuning data, but achieves competitive performance on referring expression segmentation and panoptic narrative grounding benchmarks while completely preserving LMMs’ original conversational ability. Additionally, with instructionfollowing ability preserved and grounding ability obtained, F-LMM can be directly applied to complex tasks like reasoning segmentation, grounded conversation generation and visual chain-of-thought reasoning. Our code can be found at https://github.com/wusize/F-LMM.

This study presents a new image super-resolution (SR) technique based on diffusion inversion, aiming at harnessing the rich image priors encapsulated in large pre-trained diffusion models to improve SR performance. We design a Partial noise Prediction strategy to construct an intermediate state of the diffusion model, which serves as the starting sampling point. Central to our approach is a deep noise predictor to estimate the optimal noise maps for the forward diffusion process. Once trained, this noise predictor can be used to initialize the sampling process partially along the diffusion trajectory, generating the desirable high-resolution result. Compared to existing approaches, our method offers a flexible and efficient sampling mechanism that supports an arbitrary number of sampling steps, ranging from one to five. Even with a single sampling step, our method demonstrates superior or comparable performance to recent state-of-the-art approaches. The code and model are publicly available at https://github.com/zsyOAOA/InvSR.

On top of Segment Anything Model (SAM), SAM 2 further extends its capability from image to video inputs through a memory bank mechanism and obtains a remarkable performance compared with previous methods, making it a foundation model for video segmentation task. In this paper, we aim at making SAM 2 much more efficient so that it even runs on mobile devices while maintaining a comparable performance. Despite several works optimizing SAM for better efficiency, we find they are not sufficient for SAM 2 because they all focus on compressing the image encoder, while our benchmark shows that the newly introduced memory attention blocks are also the latency bottleneck. Given this observation, we propose EdgeTAM, which leverages a novel 2D Spatial Perceiver to reduce the computational cost. In particular, the proposed 2D Spatial Perceiver encodes the densely stored frame-level memories with a lightweight Transformer that contains a fixed set of learnable queries. Given that video segmentation is a dense prediction task, we find preserving the spatial structure of the memories is essential so that the queries are split into global-level and patch-level groups. We also propose a distillation pipeline that further improves the performance without inference overhead. As a result, EdgeTAM achieves 87.7, 70.0, 72.3, and 71.7 ${\mathcal{J}}{{\& }}{\mathcal{F}}$ on DAVIS 2017, MOSE, SA-V val, and SA-V test, while running at 16 FPS on iPhone 15 Pro Max. The code and models are available here.

RGBA images, with the additional alpha channel, are crucial for any application that needs blending, masking, or transparency effects, making them more versatile than standard RGB images. Nevertheless, existing image inpainting methods are designed exclusively for RGB images. Conventional approaches to transparent image inpainting typically involve placing a background underneath RGBA images and employing a two-stage process: image inpainting followed by image matting. This pipeline, however, struggles to preserve transparency consistency in edited regions, and matting can introduce jagged edges along transparency boundaries. To address these challenges, we propose Trans-Adapter, a plug-and-play adapter that enables diffusion-based inpainting models to process transparent images directly. Trans-Adapter also supports controllable editing via ControlNet and can be seamlessly integrated into various community models. To evaluate our method, we introduce LayerBench, along with a novel non-reference alpha edge quality evaluation metric for assessing transparency edge quality. We conduct extensive experiments on LayerBench to demonstrate the effectiveness of our approach.

Photorealistic style transfer (PST) enables real-world color grading by adapting reference image colors while preserving content structure. Existing methods mainly follow either approaches: generation-based methods that prioritize stylistic fidelity at the cost of content integrity and efficiency, or global color transformation methods such as LUT, which preserve structure but lack local adaptability. To bridge this gap, we propose Spatial Adaptive 4D Look-Up Table (SA-LUT), combining LUT efficiency with neural network adaptability. SA-LUT features: (1) a Style-guided 4D LUT Generator that extracts multi-scale features from the style image to predict a 4D LUT, and (2) a Context Generator using content-style cross-attention to produce a context map. This context map enables spatially-adaptive adjustments, allowing our 4D LUT to apply precise color transformations while preserving structural integrity. To establish a rigorous evaluation framework for photorealistic style transfer, we introduce PST50, the first benchmark specifically designed for PST assessment. Experiments demonstrate that SA-LUT substantially outperforms state-of-the-art methods, achieving a 66.7% reduction in LPIPS score compared to 3D LUT approaches, while maintaining real-time performance at 16 FPS for video stylization. Our code and benchmark are available at https://github.com/Ry3nG/SA-LUT

In this work, we introduce a single parameter $ω$, to effectively control granularity in diffusion-based synthesis. This parameter is incorporated during the denoising steps of the diffusion model's reverse process. Our approach does not require model retraining, architectural modifications, or additional computational overhead during inference, yet enables precise control over the level of details in the generated outputs. Moreover, spatial masks or denoising schedules with varying $ω$ values can be applied to achieve region-specific or timestep-specific granularity control. Prior knowledge of image composition from control signals or reference images further facilitates the creation of precise $ω$ masks for granularity control on specific objects. To highlight the parameter's role in controlling subtle detail variations, the technique is named Omegance, combining "omega" and "nuance". Our method demonstrates impressive performance across various image and video synthesis tasks and is adaptable to advanced diffusion models. The code is available at https://github.com/itsmag11/Omegance.

Unifying visual understanding and generation within a single multimodal framework remains a significant challenge, as the two inherently heterogeneous tasks require representations at different levels of granularity. Current approaches that utilize vector quantization (VQ) or variational autoencoders (VAE) for unified visual representation prioritize intrinsic imagery features over semantics, compromising understanding performance. In this work, we take inspiration from masked image modelling (MIM) that learns rich semantics via a mask-and-reconstruct pre-training and its successful extension to masked autoregressive (MAR) image generation. A preliminary study on the MAR encoder's representation reveals exceptional linear probing accuracy and precise feature response to visual concepts, which indicates MAR's potential for visual understanding tasks beyond its original generation role. Based on these insights, we present \emph{Harmon}, a unified autoregressive framework that harmonizes understanding and generation tasks with a shared MAR encoder. Through a three-stage training procedure that progressively optimizes understanding and generation capabilities, Harmon achieves state-of-the-art image generation results on the GenEval, MJHQ30K and WISE benchmarks while matching the performance of methods with dedicated semantic encoders (e.g., Janus) on image understanding benchmarks. Our code and models will be available at https://github.com/wusize/Harmon.

Recent advancements in diffusion models and large-scale datasets have revolutionized image and video generation, with increasing focus on 3D content generation. While existing methods show promise, they face challenges in input formats, latent space structures, and output representations. This paper introduces a novel 3D generation framework that addresses these issues, enabling scalable and high-quality 3D generation with an interactive Point Cloud-structured Latent space. Our approach utilizes a VAE with multi-view posed RGB-D-N renderings as input, features a unique latent space design that preserves 3D shape information, and incorporates a cascaded latent flow-based model for improved shape-texture disentanglement. The proposed method, GaussianAnything, supports multi-modal conditional 3D generation, allowing for point cloud, caption, and single-view image inputs. Experimental results demonstrate superior performance on various datasets, advancing the state-of-the-art in 3D content generation.

Although learning-based image restoration methods have made significant progress, they still struggle with limited generalization to real-world scenarios due to the substantial domain gap caused by training on synthetic data. Existing methods address this issue by improving data synthesis pipelines, estimating degradation kernels, employing deep internal learning, and performing domain adaptation and regularization. Previous domain adaptation methods have sought to bridge the domain gap by learning domain-invariant knowledge in either feature or pixel space. However, these techniques often struggle to extend to low-level vision tasks within a stable and compact framework. In this paper, we show that it is possible to perform domain adaptation via the noise space using diffusion models. In particular, by leveraging the unique property of how auxiliary conditional inputs influence the multi-step denoising process, we derive a meaningful *diffusion loss* that guides the restoration model in progressively aligning both restored synthetic and real-world outputs with a target clean distribution. We refer to this method as *denoising as adaptation*. To prevent shortcuts during joint training, we present crucial strategies such as channel-shuffling layer and residual-swapping contrastive learning in the diffusion model. They implicitly blur the boundaries between conditioned synthetic and real data and prevent the reliance of the model on easily distinguishable features. Experimental results on three classical image restoration tasks, namely denoising, deblurring, and deraining, demonstrate the effectiveness of the proposed method.

Existing interpolation methods use pre‑trained video diffusion priors to generate intermediate frames between sparsely sampled keyframes. In the absence of 3D geometric guidance, these methods struggle to produce plausible results for complex, articulated human motions and offer limited control over the synthesized dynamics. In this paper, we introduce PoseFuse3D Keyframe Interpolator (PoseFuse3D-KI), a novel framework that integrates 3D human guidance signals into the diffusion process for Controllable Human-centric Keyframe Interpolation (CHKI). To provide rich spatial and structural cues for interpolation, our PoseFuse3D, a 3D‑informed control model, features a novel SMPL‑X encoder that encodes and aggregates 3D geometry and shape into the 2D latent conditioning space, alongside a fusion network that integrates these 3D cues with 2D pose embeddings. For evaluation, we build CHKI-Video, a new dataset annotated with both 2D poses and 3D SMPL‑X parameters. We show that PoseFuse3D-KI consistently outperforms state-of-the-art baselines on CHKI-Video, achieving a 9\% improvement in PSNR and a 38\% reduction in LPIPS. Comprehensive ablations demonstrate that our PoseFuse3D model improves interpolation fidelity.

In user-generated-content (UGC) applications, non-expert users often rely on image-to-3D generative models to create 3D assets. In this context, primitive-based shape abstraction offers a promising solution for UGC scenarios by compressing high-resolution meshes into compact, editable representations. Towards this end, effective shape abstraction must therefore be structure-aware, characterized by low overlap between primitives, part-aware alignment, and primitive compactness. We present Light-SQ, a novel superquadric-based optimization framework that explicitly emphasizes structure-awareness from three aspects. (a) We introduce SDF carving to iteratively udpate the target signed distance field, discouraging overlap between primitives. (b) We propose a block-regrow-fill strategy guided by structure-aware volumetric decomposition, enabling structural partitioning to drive primitive placement. (c) We implement adaptive residual pruning based on SDF update history to surpress over-segmentation and ensure compact results. In addition, Light-SQ supports multiscale fitting, enabling localized refinement to preserve fine geometric details. To evaluate our method, we introduce 3DGen-Prim, a benchmark extending 3DGen-Bench with new metrics for both reconstruction quality and primitive-level editability. Extensive experiments demonstrate that Light-SQ enables efficient, high-fidelity, and editable shape abstraction with superquadrics for complex generated geometry, advancing the feasibility of 3D UGC creation. Project Page: https://johann.wang/Light-SQ/ .

This paper proposes a large-scale multi-modal dataset for referring motion expression video segmentation, focusing on segmenting and tracking target objects in videos based on language description of objects' motions. Existing referring video segmentation datasets often focus on salient objects and use language expressions rich in static attributes, potentially allowing the target object to be identified in a single frame. Such datasets underemphasize the role of motion in both videos and languages. To explore the feasibility of using motion expressions and motion reasoning clues for pixel-level video understanding, we introduce MeViS, a dataset containing 33,072 human-annotated motion expressions in both text and audio, covering 8,171 objects in 2,006 videos of complex scenarios. We benchmark 15 existing methods across 4 tasks supported by MeViS, including 6 referring video object segmentation (RVOS) methods, 3 audio-guided video object segmentation (AVOS) methods, 2 referring multi-object tracking (RMOT) methods, and 4 video captioning methods for the newly introduced referring motion expression generation (RMEG) task. The results demonstrate weaknesses and limitations of existing methods in addressing motion expression-guided video understanding. We further analyze the challenges and propose an approach LMPM++ for RVOS/AVOS/RMOT that achieves new state-of-the-art results. Our dataset provides a platform that facilitates the development of motion expression-guided video understanding algorithms in complex video scenes.

Faithful text image super-resolution (SR) is challenging because each character has a unique structure and usually exhibits diverse font styles and layouts. While existing methods primarily focus on English text, less attention has been paid to more complex scripts like Chinese. In this paper, we introduce a high-quality text image SR framework designed to restore the precise strokes of low-resolution (LR) Chinese characters. Unlike methods that rely on character recognition priors to regularize the SR task, we propose a novel structure prior that offers structure-level guidance to enhance visual quality. Our framework incorporates this structure prior within a StyleGAN model, leveraging its generative capabilities for restoration. To maintain the integrity of character structures while accommodating various font styles and layouts, we implement a codebook-based mechanism that restricts the generative space of StyleGAN. Each code in the codebook represents the structure of a specific character, while the vector $w$w in StyleGAN controls the character's style, including typeface, orientation, and location. Through the collaborative interaction between the codebook and style, we generate a high-resolution structure prior that aligns with LR characters both spatially and structurally. Experiments demonstrate that this structure prior provides robust, character-specific guidance, enabling the accurate restoration of clear strokes in degraded characters, even for real-world LR Chinese text with irregular layouts.

While recent image warping approaches achieved remarkable success on existing benchmarks, they still require training separate models for each specific task and cannot generalize well to different camera models or customized manipulations. To address diverse types of warping in practice, we propose a Multiple-in-One image WArping model (named MOWA) in this work. Specifically, we mitigate the difficulty of multi-task learning by disentangling the motion estimation at both the region level and pixel level. To further enable dynamic task-aware image warping, we introduce a lightweight point-based classifier that predicts the task type, serving as prompts to modulate the feature maps for more accurate estimation. To our knowledge, this is the first work that solves multiple practical warping tasks in one single model. Extensive experiments demonstrate that our MOWA, which is trained on six tasks for multiple-in-one single image warping, outperforms state-of-the-art task-specific models across most tasks. Moreover, MOWA also exhibits promising potential to generalize into unseen scenes, as evidenced by cross-domain and zero-shot evaluations.

While diffusion-based image restoration (IR) methods have achieved remarkable success, they are still limited by the low inference speed attributed to the necessity of executing hundreds or even thousands of sampling steps. Existing acceleration sampling techniques, though seeking to expedite the process, inevitably sacrifice performance to some extent, resulting in over-blurry restored outcomes. To address this issue, this study proposes a novel and efficient diffusion model for IR that significantly reduces the required number of diffusion steps. Our method avoids the need for post-acceleration during inference, thereby avoiding the associated performance deterioration. Specifically, our proposed method establishes a Markov chain that facilitates the transitions between the high-quality and low-quality images by shifting their residuals, substantially improving the transition efficiency. A carefully formulated noise schedule is devised to flexibly control the shifting speed and the noise strength during the diffusion process. Extensive experimental evaluations demonstrate that the proposed method achieves superior or comparable performance to current state-of-the-art methods on four classical IR tasks, namely image super-resolution, image inpainting, blind face restoration, and image deblurring, even only with four sampling steps.

Open-vocabulary object detection (OVOD) aims to detect the objects beyond the set of classes observed during training. This work introduces a straightforward and efficient strategy that utilizes pre-trained vision-language models (VLM), like CLIP, to identify potential novel classes through zero-shot classification. Previous methods use a class-agnostic region proposal network to detect object proposals and consider the proposals that do not match the ground truth as background. Unlike these methods, our method selects a subset of background proposals and treats them as novel classes during training. We refer to this approach as the self-training strategy, which enhances recall and accuracy for novel classes without requiring extra annotations, datasets, and re-training. Compared to previous pseudo methods, our approach does not require re-training and offline labeling processing, which is more efficient and effective in one-shot training. Empirical evaluations on three datasets, including LVIS, V3Det, and COCO, demonstrate significant improvements over the baseline performance without incurring additional parameters or computational costs during inference. In addition, we also apply our method to various baselines. In particular, compared with the previous method, F-VLM, our method achieves a 2.5% improvement on the LVIS dataset. Combined with the recent method CLIPSelf, our method also achieves 46.7 novel class AP on COCO without introducing extra data for pertaining. We also achieve over 6.5% improvement over the F-VLM baseline in the recent challenging V3Det dataset. We release our code and models athttps://github.com/xushilin1/dst-det.

Detecting objects accurately from a large or open vocabulary necessitates the vision-language alignment on region representations. However, learning such a region-text alignment by obtaining high-quality box annotations with text labels or descriptions is expensive and infeasible. In contrast, collecting image-text pairs is simpler but lacks precise object location information to associate regions with texts. In this paper, we propose a novel approach called Contrastive Language-Image Mosaic (CLIM), which leverages large-scale image-text pairs effectively for aligning region and text representations. CLIM combines multiple images into a mosaicked image and treats each image as a ‘pseudo region’. The feature of each pseudo region is extracted and trained to be similar to the corresponding text embedding while dissimilar from others by a contrastive loss, enabling the model to learn the region-text alignment without costly box annotations. As a generally applicable approach, CLIM consistently improves different open-vocabulary object detection methods that use caption supervision. Furthermore, CLIM can effectively enhance the region representation of vision-language models, thus providing stronger backbones for open-vocabulary object detectors. Our experimental results demonstrate that CLIM improves different baseline open-vocabulary object detectors by a large margin on both OV-COCO and OV-LVIS benchmarks. The code is available at https://github.com/wusize/CLIM.

Recent advances in zero-shot text-to-3D human generation, which employ the human model prior (e.g., SMPL) or Score Distillation Sampling (SDS) with pre-trained text-to-image diffusion models, have been groundbreaking. However, SDS may provide inaccurate gradient directions under the weak diffusion guidance, as it tends to produce over-smoothed results and generate body textures that are inconsistent with the detailed mesh geometry. Therefore, directly leveraging existing strategies for high-fidelity text-to-3D human texturing is challenging. In this work, we propose a model called PaintHuman to addresses the challenges from two perspectives. We first propose a novel score function, Denoised Score Distillation (DSD), which directly modifies the SDS by introducing negative gradient components to iteratively correct the gradient direction and generate high-quality textures. In addition, we use the depth map as a geometric guide to ensure that the texture is semantically aligned to human mesh surfaces. To guarantee the quality of rendered results, we employ geometry-aware networks to predict surface materials and render realistic human textures. Extensive experiments, benchmarked against state-of-the-art (SoTA) methods, validate the efficacy of our approach.Project page: https://painthuman.github.io/.

Text-to-image diffusion models have remarkably excelled in producing diverse, high-quality, and photo-realistic images. This advancement has spurred a growing interest in incorporating specific identities into generated content. Most current methods employ an inversion approach to em-bed a target visual concept into the text embedding space using a single reference image. However, the newly synthe-sized faces either closely resemble the reference image in terms of facial attributes, such as expression, or exhibit a reduced capacity for identity preservation. Text descriptions intended to guide the facial attributes of the synthesized face may fall short, owing to the intricate entanglement of identity information with identity-irrelevant facial attributes derived from the reference image. To address these issues, we present the novel use of the extended StyleGAN embed-ding space$\mathcal{W}+$, to achieve enhanced identity preservation and disentanglement for diffusion models. By aligning this semantically meaningful human face latent space with text-to-image diffusion models, we succeed in maintaining high fidelity in identity preservation, coupled with the capacity for semantic editing. Additionally, we propose new training objectives to balance the influences of both prompt and identity conditions, ensuring that the identity-irrelevant back-ground remains negligibly affected during facial attribute modifications. Extensive experiments reveal that our method adeptly generates personalized text-to-image outputs that are not only compatible with prompt descriptions but also amenable to common StyleGAN editing directions in diverse settings. Our code and model are available at https://github.com/csxmli2016/w-plus-adapter.

Text-driven video generation witnesses rapid progress. However, merely using text prompts is not enough to depict the desired subject appearance that accurately aligns with users' intents, especially for customized content creation. In this paper, we study the task of video generation with image prompts, which provide more accurate and direct content control beyond the text prompts. Specifically, we propose a feed-forward framework VideoBooth, with two dedicated designs: 1) We propose to embed image prompts in a coarse-to-fine manner. Coarse visual embeddings from image encoder provide high-level encodings of image prompts, while fine visual embeddings from the proposed attention injection module provide multi-scale and detailed encoding of image prompts. These two complementary embeddings can faithfully capture the desired appearance. 2) In the attention injection module at fine level, multi-scale image prompts are fed into different cross-frame attention layers as additional keys and values. This extra spatial in-formation refines the details in the first frame and then it is propagated to the remaining frames, which maintains temporal consistency. Extensive experiments demonstrate that Video Booth achieves state-of-the-art performance in gener-ating customized high-quality videos with subjects specified in image prompts. Notably, VideoBooth is a generalizable framework where a single model works for a wide range of image prompts with only feed-forward passes.

In this work, we present Digital Life Project, a framework utilizing language as the universal medium to build autonomous 3D characters, who are capable of engaging in social interactions and expressing with articulated body motions, thereby simulating life in a digital environment. Our framework comprises two primary components: 1) SocioMind: a meticulously crafted digital brain that models personalities with systematic few-shot exemplars, incorporates a reflection process based on psychology principles, and emulates autonomy by initiating dialogue topics; 2) MoMat-MoGen: a text-driven motion synthesis paradigm for controlling the character's digital body. It integrates motion matching, a proven industry technique to ensure motion quality, with cutting-edge advancements in motion generation for diversity. Extensive experiments demonstrate that each module achieves state-of-the-art performance in its respective domain. Collectively, they enable virtual characters to initiate and sustain dialogues autonomously, while evolving their socio-psychological states. Concurrently, these characters can perform contextually relevant bodily movements. Additionally, an extension of DLP enables a virtual character to recognize and appropriately respond to human players' actions.

Despite the emergence of successful NeRF inpainting methods built upon explicit RGB and depth 2D inpainting supervisions, these methods are inherently constrained by the capabilities of their underlying 2D inpainters. This is due to two key reasons: (i) independently inpainting constituent images results in view-inconsistent imagery, and (ii) 2D inpainters struggle to ensure high-quality geometry completion and alignment with inpainted RGB images. To overcome these limitations, we propose a novel approach called MVIP-NeRF that harnesses the potential of diffusion priors for NeRF inpainting, addressing both appearance and geometry aspects. MVIP-NeRF performs joint inpainting across multiple views to reach a consistent solution, which is achieved via an iterative optimization process based on Score Distillation Sampling (SDS). Apart from recovering the rendered RGB images, we also extract normal maps as a geometric representation and define a normal SDS loss that motivates accurate geometry inpaint- ing and alignment with the appearance. Additionally, we formulate a multi-view SDS score function to distill generative priors simultaneously from different view images, ensuring consistent visual completion when dealing with large view variations. Our experimental results show better appearance and geometry recovery than previous NeRF in painting methods.

Colorizing line art is a pivotal task in the production of hand-drawn cel animation. This typically involves digital painters using a paint bucket tool to manually color each segment enclosed by lines, based on RGB values predeter-mined by a color designer. This frame-by-frame process is both arduous and time-intensive. Current automated meth-ods mainly focus on segment matching. This technique mi-grates colors from a reference to the target frame by aligning features within line-enclosed segments across frames. However, issues like occlusion and wrinkles in animations often disrupt these direct correspondences, leading to mis-matches. In this work, we introduce a new learning-based inclusion matching pipeline, which directs the network to comprehend the inclusion relationships between segments rather than relying solely on direct visual correspondences. Our method features a two-stage pipeline that integrates a coarse color warping module with an inclusion matching module, enabling more nuanced and accurate colorization. To facilitate the training of our network, we also develope a unique dataset, referred to as PaintBucket-Character. This dataset includes rendered line arts alongside their colorized counterparts, featuring various 3D characters. Extensive experiments demonstrate the effectiveness and superiority of our method over existing techniques.

In this work, we address various segmentation tasks, each traditionally tackled by distinct or partially unified models. We propose OMG-Seg, One Model that is Good enough to efficiently and effectively handle all the segmentation tasks, including image semantic, instance, and panoptic segmentation, as well as their video counterparts, open vocabulary settings, prompt-driven, interactive segmentation like SAM, and video object segmentation. To our knowledge, this is the first model to handle all these tasks in one model and achieve satisfactory performance. We show that OMG-Seg, a transformer-based encoder-decoder architecture with task-specific queries and outputs, can support over ten distinct segmentation tasks and yet significantly reduce computational and parameter overhead across various tasks and datasets. We rigorously evaluate the inter-task influences and correlations during co-training. Code and models are available at https://github.com/lxtGH/OMG-Seg.

We introduce a new task - language-driven video inpainting, which uses natural language instructions to guide the inpainting process. This approach overcomes the limitations of traditional video inpainting methods that depend on manually labeled binary masks, a process often tedious and labor-intensive. We present the Remove Objects from Videos by Instructions (ROVI) dataset, containing 5,650 videos and 9,091 inpainting results, to support training and evaluation for this task. We also propose a novel diffusion-based language-driven video inpainting framework, the first end-to-end baseline for this task, integrating Multimodal Large Language Models to understand and execute complex language-based inpaintingrequests effectively. Our comprehensive results showcase the dataset's versatility and the model's effectiveness in various language-instructed inpainting scenarios. We have made datasets, code, and models publicly available at https://github.com/jianzongwu/Language-Driven-Video-Inpainting.

The remarkable efficacy of text-to-image diffusion models has motivated extensive exploration of their potential application in video domains. Zero-shot methods seek to extend image diffusion models to videos without necessitating model training. Recent methods mainly focus on incorporating inter-frame correspondence into attention mechanisms. However, the soft constraint imposed on determining where to attend to valid features can sometimes be insufficient, resulting in temporal inconsistency. In this paper, we introduce FRESCO, intra-frame correspondence alongside inter-frame correspondence to establish a more robust spatial-temporal constraint. This enhancement ensures a more consistent transformation of semantically similar content across frames. Beyond mere attention guidance, our approach involves an explicit update of features to achieve high spatial-temporal consistency with the input video, significantly improving the visual coherence of the resulting translated videos. Extensive experiments demonstrate the effectiveness of our proposed framework in producing high-quality, coherent videos, marking a notable improvement over existing zero-shot methods.

Text-based diffusion models have exhibited remarkable success in generation and editing, showing great promise for enhancing visual content with their generative prior. However, applying these models to video super-resolution remains challenging due to the high demands for output fidelity and temporal consistency, which is complicated by the inherent randomness in diffusion models. Our study introduces Upscale-A-Video, a text-guided latent diffusion framework for video upscaling. This framework ensures temporal coherence through two key mechanisms: locally, it integrates temporal layers into U-Net and VAE-Decoder, maintaining consistency within short sequences; globally, without training, a flow-guided recurrent latent propagation module is introduced to enhance overall video stability by propagating and fusing latent across the entire sequences. Thanks to the diffusion paradigm, our model also offers greater flexibility by allowing text prompts to guide texture creation and adjustable noise levels to balance restoration and generation, enabling a trade-off between fidelity and quality. Extensive experiments show that Upscale-A-Video surpasses existing methods in both synthetic and real-world benchmarks, as well as in AI-generated videos, showcasing impressive visual realism and temporal consistency.

Local motion blur commonly occurs in real-world photography due to the mixing between moving objects and stationary backgrounds during exposure. Existing image deblurring methods predominantly focus on global deblurring, inadvertently affecting the sharpness of backgrounds in locally blurred images and wasting unnecessary computation on sharp pixels, especially for high-resolution images. This paper aims to adaptively and efficiently restore high-resolution locally blurred images. We propose a local motion deblurring vision Transformer (LMD-ViT) built on adaptive window pruning Transformer blocks (AdaWPT). To focus deblurring on local regions and reduce computation, AdaWPT prunes unnecessary windows, only allowing the active windows to be involved in the deblurring processes. The pruning operation relies on the blurriness confidence predicted by a confidence predictor that is trained end-to-end using a reconstruction loss with Gumbel-Softmax re-parameterization and a pruning loss guided by annotated blur masks. Our method removes local motion blur effectively without distorting sharp regions, demonstrated by its exceptional perceptual and quantitative improvements (+0.28dB) compared to state-of-the-art methods. In addition, our approach substantially reduces FLOPs by 66% and achieves more than a twofold increase in inference speed compared to Transformer-based deblurring methods. We will make our code and annotated blur masks publicly available.

We introduce a novel task within the field of human motion generation, termed dance accompaniment, which necessitates the generation of responsive movements from a dance partner, the "follower", synchronized with the lead dancer’s movements and the underlying musical rhythm. Unlike existing solo or group dance generation tasks, a duet dance scenario entails a heightened degree of interaction between the two participants, requiring delicate coordination in both pose and position. To support this task, we first build a large-scale and diverse duet interactive dance dataset, DD100, by recording about 117 minutes of professional dancers’ performances. To address the challenges inherent in this task, we propose a GPT based model, Duolando, which autoregressively predicts the subsequent tokenized motion conditioned on the coordinated information of the music, the leader’s and the follower’s movements. To further enhance the GPT’s capabilities of generating stable results on unseen conditions (music and leader motions), we devise an off-policy reinforcement learning strategy that allows the model to explore viable trajectories from out-of-distribution samplings, guided by human-defined rewards. Based on the collected dataset and proposed method, we establish a benchmark with several carefully designed metrics.

Open-vocabulary dense prediction tasks including object detection and image segmentation have been advanced by the success of Contrastive Language-Image Pre-training (CLIP). CLIP models, particularly those incorporating vision transformers (ViTs), have exhibited remarkable generalization ability in zero-shot image classification. However, when transferring the vision-language alignment of CLIP from global image representation to local region representation for the open-vocabulary dense prediction tasks, CLIP ViTs suffer from the domain shift from full images to local image regions. In this paper, we embark on an in-depth analysis of the region-language alignment in CLIP models, which is essential for downstream open-vocabulary dense prediction tasks. Subsequently, we propose an approach named CLIPSelf, which adapts the image-level recognition ability of CLIP ViT to local image regions without needing any region-text pairs. CLIPSelf empowers ViTs to distill itself by aligning a region representation extracted from its dense feature map with the image-level representation of the corresponding image crop. With the enhanced CLIP ViTs, we achieve new state-of-the-art performance on open-vocabulary object detection, semantic segmentation, and panoptic segmentation across various benchmarks. Models and code are released at https://github.com/wusize/CLIPSelf.

Motion modeling is critical in flow-based Video Frame Interpolation (VFI). Existing paradigms either consider linear combinations of bidirectional flows or directly predict bilateral flows for given timestamps without exploring favorable motion priors, thus lacking the capability of effectively modeling spatiotemporal dynamics in real-world videos. To address this limitation, in this study, we introduce Generalizable Implicit Motion Modeling (GIMM), a novel and effective approach to motion modeling for VFI. Specifically, to enable GIMM as an effective motion modeling paradigm, we design a motion encoding pipeline to model spatiotemporal motion latent from bidirectional flows extracted from pre-trained flow estimators, effectively representing input-specific motion priors. Then, we implicitly predict arbitrary-timestep optical flows within two adjacent input frames via an adaptive coordinate-based neural network, with spatiotemporal coordinates and motion latent as inputs. Our GIMM can be easily integrated with existing flow-based VFI works by supplying accurately modeled motion. We show that GIMM performs better than the current state of the art on standard VFI benchmarks.

Garment animation is ubiquitous in various applications, such as virtual reality, gaming, and film producing. Recently, learning-based approaches obtain compelling performance in animating diverse garments under versatile scenarios. Nevertheless, to mimic the deformations of the observed garments, data-driven methods require large scale of garment data, which are both resource-wise expensive and time-consuming. In addition, forcing models to match the dynamics of observed garment animation may hinder the potentials to generalize to unseen cases. In this paper, instead of using garment-wise supervised-learning we adopt a disentangled scheme to learn how to animate observed garments: 1). learning constitutive behaviors from the observed cloth; 2). dynamically animate various garments constrained by the learned constitutive laws. Specifically, we propose Energy Unit network (EUNet) to model the constitutive relations in the format of energy. Without the priors from analytical physics models and differentiable simulation engines, EUNet is able to directly capture the constitutive behaviors from the observed piece of cloth and uniformly describes the change of energy caused by deformations, such as stretching and bending. We further apply the pre-trained EUNet to animate various garments based on energy optimizations. The disentangled scheme alleviates the need of garment data and enables us to utilize the dynamics of a piece of cloth for animating garments. Experiments show that while EUNet effectively delivers the energy gradients due to the deformations, models constrained by EUNet achieve more stable and physically plausible performance comparing with those trained in garment-wise supervised manner.

Current universal segmentation methods demonstrate strong capabilities in pixel-level image and video understanding. However, they lack reasoning abilities and cannot be controlled via text instructions. In contrast, large vision-language multimodal models exhibit powerful vision-based conversation and reasoning capabilities but lack pixel-level understanding and have difficulty accepting visual prompts for flexible user interaction. This paper proposes OMG-LLaVA, a new and elegant framework combining powerful pixel-level vision understanding with reasoning abilities. It can accept various visual and text prompts for flexible user interaction. Specifically, we use a universal segmentation method as the visual encoder, integrating image information, perception priors, and visual prompts into visual tokens provided to the LLM. The LLM is responsible for understanding the user's text instructions and providing text responses and pixel-level segmentation results based on the visual information. We propose perception prior embedding to better integrate perception priors with image features. OMG-LLaVA achieves image-level, object-level, and pixel-level reasoning and understanding in a single model, matching or surpassing the performance of specialized methods on multiple benchmarks. Rather than using LLM to connect each specialist, our work aims at end-to-end training on one encoder, one decoder, and one LLM. The code and model have been released for further research.

Digital human motion synthesis is a vibrant research field with applications in movies, AR/VR, and video games. Whereas methods were proposed to generate natural and realistic human motions, most only focus on modeling humans and largely ignore object movements. Generating task-oriented human-object interaction motions in simulation is challenging. For different intents of using the objects, humans conduct various motions, which requires the human first to approach the objects and then make them move consistently with the human instead of staying still. Also, to deploy in downstream applications, the synthesized motions are desired to be flexible in length, providing options to personalize the predicted motions for various purposes. To this end, we propose TOHO: Task-Oriented Human-Object Interactions Generation with Implicit Neural Representations, which generates full human-object interaction motions to conduct specific tasks, given only the task type, the object, and a starting human status. TOHO generates human-object motions in four steps: 1) it first estimates the object’s final position given the task intent; 2) it then generates keyframe poses grasping the objects; 3) after that, it infills the keyframes and generates continuous motions; 4) finally, it applies a compact closed-form object motion estimation to generate the object motion. Our method generates continuous motions that are parameterized only by the temporal coordinate, which allows for upsampling of the sequence to arbitrary frames and adjusting the motion speeds by designing the temporal coordinate vector. This work takes a step further toward general human-scene interaction simulation.

Video colorization is a challenging and highly ill-posed problem. Although recent years have witnessed remarkable progress in single image colorization, there is relatively less research effort on video colorization, and existing methods always suffer from severe flickering artifacts (temporal inconsistency) or unsatisfactory colorization. We address this problem from a new perspective, by jointly considering colorization and temporal consistency in a unified framework. Specifically, we propose a novel temporally consistent video colorization (TCVC) framework. TCVC effectively propagates frame-level deep features in a bidirectional way to enhance the temporal consistency of colorization. Furthermore, TCVC introduces a self-regularization learning (SRL) scheme to minimize the differences in predictions obtained using different time steps. SRL does not require any ground-truth color videos for training and can further improve temporal consistency. Experiments demonstrate that our method can not only provide visually pleasing colorized video, but also with clearly better temporal consistency than state-of-the-art methods. A video demo is provided at https://www.youtube.com/watch?v=c7dczMs-olE , while code is available at https://github.com/lyh-18/TCVC-Temporally-Consistent-Video-Colorization .

Image- and video-based 3D human recovery (i.e., pose and shape estimation) have achieved substantial progress. However, due to the prohibitive cost of motion capture, existing datasets are often limited in scale and diversity. In this work, we obtain massive human sequences by playing the video game with automatically annotated 3D ground truths. Specifically, we contribute GTA-Human, a large-scale 3D human dataset generated with the GTA-V game engine, featuring a highly diverse set of subjects, actions, and scenarios. More importantly, we study the use of game-playing data and obtain five major insights. First, game-playing data is surprisingly effective. A simple frame-based baseline trained on GTA-Human outperforms more sophisticated methods by a large margin. For video-based methods, GTA-Human is even on par with the in-domain training set. Second, we discover that synthetic data provides critical complements to the real data that is typically collected indoor. We highlight that our investigation into domain gap provides explanations for our data mixture strategies that are simple yet useful, which offers new insights to the research community. Third, the scale of the dataset matters. The performance boost is closely related to the additional data available. A systematic study on multiple key factors (such as camera angle and body pose) reveals that the model performance is sensitive to data density. Fourth, the effectiveness of GTA-Human is also attributed to the rich collection of strong supervision labels (SMPL parameters), which are otherwise expensive to acquire in real datasets. Fifth, the benefits of synthetic data extend to larger models such as deeper convolutional neural networks (CNNs) and Transformers, for which a significant impact is also observed. We hope our work could pave the way for scaling up 3D human recovery to the real world.

Artificial lights commonly leave strong lens flare artifacts on the images captured at night, degrading both the visual quality and performance of vision algorithms. Existing flare removal approaches mainly focus on removing daytime flares and fail in nighttime cases. Nighttime flare removal is challenging due to the unique luminance and spectrum of artificial lights, as well as the diverse patterns and image degradation of the flares. The scarcity of the nighttime flare removal dataset constrains the research on this crucial task. In this paper, we introduce Flare7K++, the first comprehensive nighttime flare removal dataset, consisting of 962 real-captured flare images (Flare-R) and 7000 synthetic flares (Flare7K). Compared to Flare7K, Flare7K++ is particularly effective in eliminating complicated degradation around the light source, which is intractable by using synthetic flares alone. Besides, the previous flare removal pipeline relies on the manual threshold and blur kernel settings to extract light sources, which may fail when the light sources are tiny or not overexposed. To address this issue, we additionally provide the annotations of light sources in Flare7K++ and propose a new end-to-end pipeline to preserve the light source while removing lens flares. Our dataset and pipeline offer a valuable foundation and benchmark for future investigations into nighttime flare removal studies. Extensive experiments demonstrate that Flare7K++ supplements the diversity of existing flare datasets and pushes the frontier of nighttime flare removal toward real-world scenarios.

Facial editing is to manipulate the facial attributes of a given face image. Nowadays, with the development of generative models, users can easily generate 2D and 3D facial images with high fidelity and 3D-aware consistency. However, existing works are incapable of delivering a continuous and fine-grained editing mode (e.g., editing a slightly smiling face to a big laughing one) with natural interactions with users. In this work, we propose Talk-to-Edit, an interactive facial editing framework that performs fine-grained attribute manipulation through dialog between the user and the system. Our key insight is to model a continual "semantic field" in the GAN latent space. 1) Unlike previous works that regard the editing as traversing straight lines in the latent space, here the fine-grained editing is formulated as finding a curving trajectory that respects fine-grained attribute landscape on the semantic field. 2) The curvature at each step is location-specific and determined by the input image as well as the users' language requests. 3) To engage the users in a meaningful dialog, our system generates language feedback by considering both the user request and the current state of the semantic field. We demonstrate the effectiveness of our proposed framework on both 2D and 3D-aware generative models. We term the semantic field for the 3D-aware models as "tri-plane" flow, as it corresponds to the changes not only in the color space but also in the density space. We also contribute CelebA-Dialog, a visual-language facial editing dataset to facilitate large-scale study. Specifically, each image has manually annotated fine-grained attribute annotations as well as template-based textual descriptions in natural language. Extensive quantitative and qualitative experiments demonstrate the superiority of our framework in terms of 1) the smoothness of fine-grained editing, 2) the identity/attribute preservation, and 3) the visual photorealism and dialog fluency. Notably, the user study validates that our overall system is consistently favored by around 80% of the participants.

Visual segmentation seeks to partition images, video frames, or point clouds into multiple segments or groups. This technique has numerous real-world applications, such as autonomous driving, image editing, robot sensing, and medical analysis. Over the past decade, deep learning-based methods have made remarkable strides in this area. Recently, transformers, a type of neural network based on self-attention originally designed for natural language processing, have considerably surpassed previous convolutional or recurrent approaches in various vision processing tasks. Specifically, vision transformers offer robust, unified, and even simpler solutions for various segmentation tasks. This survey provides a thorough overview of transformer-based visual segmentation, summarizing recent advancements. We first review the background, encompassing problem definitions, datasets, and prior convolutional methods. Next, we summarize a meta-architecture that unifies all recent transformer-based approaches. Based on this meta-architecture, we examine various method designs, including modifications to the meta-architecture and associated applications. We also present several specific subfields, including 3D point cloud segmentation, foundation model tuning, domain-aware segmentation, efficient segmentation, and medical segmentation. Additionally, we compile and re-evaluate the reviewed methods on several well-established datasets. Finally, we identify open challenges in this field and propose directions for future research.

Neural Radiance Field (NeRF) has achieved substantial progress in novel view synthesis given multi-view images. Recently, some works have attempted to train a NeRF from a single image with 3D priors. They mainly focus on a limited field of view with a few occlusions, which greatly limits their scalability to real-world 360-degree panoramic scenarios with large-size occlusions. In this paper, we present PERF, a 360-degree novel view synthesis framework that trains a panoramic neural radiance field from a single panorama. Notably, PERF allows 3D roaming in a complex scene without expensive and tedious image collection. To achieve this goal, we propose a novel collaborative RGBD inpainting method and a progressive inpainting-and-erasing method to lift up a 360-degree 2D scene to a 3D scene. Specifically, we first predict a panoramic depth map as initialization given a single panorama and reconstruct visible 3D regions with volume rendering. Then we introduce a collaborative RGBD inpainting approach into a NeRF for completing RGB images and depth maps from random views, which is derived from an RGB Stable Diffusion model and a monocular depth estimator. Finally, we introduce an inpainting-and-erasing strategy to avoid inconsistent geometry between a newly-sampled view and reference views. The two components are integrated into the learning of NeRFs in a unified optimization framework and achieve promising results. Extensive experiments on Replica and a new dataset PERF-in-the-wild demonstrate the superiority of our PERF over state-of-the-art methods. Our PERF can be widely used for real-world applications, such as panorama-to-3D, text-to-3D, and 3D scene stylization applications.

While deep learning-based methods for blind face restoration have achieved unprecedented success, they still suffer from two major limitations. First, most of them deteriorate when facing complex degradations out of their training data. Second, these methods require multiple constraints, e.g., fidelity, perceptual, and adversarial losses, which require laborious hyper-parameter tuning to stabilize and balance their influences. In this work, we propose a novel method namedDifFacethat is capable of coping with unseen and complex degradations more gracefully without complicated loss designs. The key of our method is to establish a posterior distribution from the observed low-quality (LQ) image to its high-quality (HQ) counterpart. In particular, we design a transition distribution from the LQ image to the intermediate state of a pre-trained diffusion model and then gradually transmit from this intermediate state to the HQ target by recursively applying a pre-trained diffusion model. The transition distribution only relies on a restoration backbone that is trained with$L_{1}$loss on some synthetic data, which favorably avoids the cumbersome training process in existing methods. Moreover, the transition distribution can contract the error of the restoration backbone and thus makes our method more robust to unknown degradations. Comprehensive experiments show thatDifFaceis superior to current state-of-the-art methods, especially in cases with severe degradations.

Measuring the perception of visual content is a long-standing problem in computer vision. Many mathematical models have been developed to evaluate the look or quality of an image. Despite the effectiveness of such tools in quantifying degradations such as noise and blurriness levels, such quantification is loosely coupled with human language. When it comes to more abstract perception about the feel of visual content, existing methods can only rely on supervised models that are explicitly trained with labeled data collected via laborious user study. In this paper, we go beyond the conventional paradigms by exploring the rich visual language prior encapsulated in Contrastive Language-Image Pre-training (CLIP) models for assessing both the quality perception (look) and abstract perception (feel) of images without explicit task-specific training. In particular, we discuss effective prompt designs and show an effective prompt pairing strategy to harness the prior. We also provide extensive experiments on controlled datasets and Image Quality Assessment (IQA) benchmarks. Our results show that CLIP captures meaningful priors that generalize well to different perceptual assessments.

Blind text image super-resolution (SR) is challenging as one needs to cope with diverse font styles and unknown degradation. To address the problem, existing methods perform character recognition in parallel to regularize the SR task, either through a loss constraint or intermediate feature condition. Nonetheless, the high-level prior could still fail when encountering severe degradation. The prob-lem is further compounded given characters of complex structures, e.g., Chinese characters that combine multiple pictographic or ideographic symbols into a single charac-ter. In this work, we present a novel prior that focuses more on the character structure. In particular, we learn to encapsulate rich and diverse structures in a StyleGAN and exploit such generative structure priors for restoration. To restrict the generative space of StyleGAN so that it obeys the structure of characters yet remains flexible in handling different font styles, we store the discrete features for each character in a codebook. The code subsequently drives the StyleGAN to generate high-resolution structural details to aid text SR. Compared to priors based on character recognition, the proposed structure prior ex-erts stronger character-specific guidance to restore faithful and precise strokes of a designated character. Extensive experiments on synthetic and real datasets demonstrate the compelling performance of the proposed generative structure prior in facilitating robust text SR. Our code is available at https://github.com/csxmli2016/MARCONet.

Recent self-supervised methods are mainly designed for representation learning with the base model, e.g., ResNets or ViTs. They cannot be easily transferred to DETR, with task-specific Transformer modules. In this work, we present Siamese DETR, a Siamese self-supervised pretraining approach for the Transformer architecture in DETR. We consider learning view-invariant and detection-oriented representations simultaneously through two complementary tasks, i.e., localization and discrimination, in a novel multi-view learning framework. Two self-supervised pretext tasks are designed: (i) Multi-View Region Detection aims at learning to localize regions-of-interest between augmented views of the input, and (ii) Multi-View Semantic Discrimination attempts to improve object-level discrimination for each region. The proposed Siamese DETR achieves state-of-the-art transfer performance on COCO and PASCAL VOC detection using different DETR variants in all setups. Code is available at https://github.com/Zx55/SiameseDEtr.

Reflective flare is a phenomenon that occurs when light reflects inside lenses, causing bright spots or a “ghosting effect” in photos, which can impact their quality. Eliminating reflective flare is highly desirable but challenging. Many existing methods rely on manually designed features to detect these bright spots, but they often fail to identify reflective flares created by various types of light and may even mistakenly remove the light sources in scenarios with multiple light sources. To address these challenges, we propose an optical center symmetry prior, which suggests that the reflective flare and light source are always symmetrical around the lens's optical center. This prior helps to locate the reflective flare's proposal region more accurately and can be applied to most smartphone cameras. Building on this prior, we create the first reflective flare removal dataset called BracketFlare, which contains diverse and realistic reflective flare patterns. We use continuous bracketing to capture the reflective flare pattern in the underexposed image and combine it with a normally exposed image to synthesize a pair of flare-corrupted and flare-free images. With the dataset, neural networks can be trained to remove the reflective flares effectively. Extensive experiments demonstrate the effectiveness of our method on both synthetic and real-world datasets.

Due to the difficulty in collecting large-scale and perfectly aligned paired training data for Under-Display Camera (UDC) image restoration, previous methods resort to monitor-based image systems or simulation-based methods, sacrificing the realness of the data and introducing domain gaps. In this work, we revisit the classic stereo setup for training data collection – capturing two images of the same scene with one UDC and one standard camera. The key idea is to “copy” details from a high-quality reference image and “paste” them on the UDC image. While being able to generate real training pairs, this setting is susceptible to spatial misalignment due to perspective and depth of field changes. The problem is further compounded by the large domain discrepancy between the UDC and normal images, which is unique to UDC restoration. In this paper, we mitigate the non-trivial domain discrepancy and spatial misalignment through a novel Transformer-based framework that generates well-aligned yet high-quality target data for the corresponding UDC input. This is made possible through two carefully designed components, namely, the Domain Alignment Module (DAM) and Geometric Alignment Module (GAM), which encourage robust and accurate discovery of correspondence between the UDC and normal views. Extensive experiments show that high-quality and well-aligned pseudo UDC training pairs are beneficial for training a robust restoration network. Code and the dataset are available at https://github.com/jnjaby/AlignFormer.

StyleGAN has achieved great progress in 2D face reconstruction and semantic editing via image inversion and latent editing. While studies over extending 2D StyleGAN to 3D faces have emerged, a corresponding generic 3D GAN inversion framework is still missing, limiting the applications of 3D face reconstruction and semantic editing. In this paper, we study the challenging problem of 3D GAN inversion where a latent code is predicted given a single face image to faithfully recover its 3D shapes and detailed textures. The problem is ill-posed: innumerable compositions of shape and texture could be rendered to the current image. Furthermore, with the limited capacity of a global latent code, 2D inversion methods cannot preserve faithful shape and texture at the same time when applied to 3D models. To solve this problem, we devise an effective self-training scheme to constrain the learning of inversion. The learning is done efficiently without any real-world 2D-3D training pairs but proxy samples generated from a 3D GAN. In addition, apart from a global latent code that captures the coarse shape and texture information, we augment the generation network with a local branch, where pixel-aligned features are added to faithfully reconstruct face details. We further consider a new pipeline to perform 3D view-consistent editing. Extensive experiments show that our method outperforms state-of-the-art inversion methods in both shape and texture reconstruction quality.

We introduce Correlational Image Modeling (CIM), a novel and surprisingly effective approach to self-supervised visual pre-training. Our CIM performs a simple pretext task: we randomly crop image regions (exemplars) from an input image (context) and predict correlation maps between the exemplars and the context. Three key designs enable correlational image modeling as a nontrivial and meaningful self-supervisory task. First, to generate useful exemplar-context pairs, we consider cropping image regions with various scales, shapes, rotations, and transformations. Second, we employ a bootstrap learning framework that involves online and target encoders. During pre-training, the former takes exemplars as inputs while the latter converts the context. Third, we model the output correlation maps via a simple cross-attention block, within which the context serves as queries and the exemplars offer values and keys. We show that CIM performs on par or better than the current state of the art on self-supervised and transfer benchmarks. Code is available at https://github.com/weivision/Correlational-Image-Modeling.git.

Pre-trained vision-language models (VLMs) learn to align vision and language representations on large-scale datasets, where each image-text pair usually contains a bag of semantic concepts. However, existing open-vocabulary object detectors only align region embeddings individually with the corresponding features extracted from the VLMs. Such a design leaves the compositional structure of semantic concepts in a scene under-exploited, although the structure may be implicitly learned by the VLMs. In this work, we propose to align the embedding of bag of regions beyond individual regions. The proposed method groups contextually interrelated regions as a bag. The embeddings of regions in a bag are treated as embeddings of words in a sentence, and they are sent to the text encoder of a VLM to obtain the bag-of-regions embedding, which is learned to be aligned to the corresponding features extracted by a frozen VLM. Applied to the commonly used Faster R-CNN, our approach surpasses the previous best results by 4.6 box AP50and 2.8 mask AP on novel categories of open-vocabulary COCO and LVIS benchmarks, respectively. Code and models are available at https://github.com/wusize/ovdet.

Towards building comprehensive real-world visual perception systems, we propose and study a new problem called panoptic scene graph generation (PVSG). PVSG is related to the existing video scene graph generation (VidSGG) problem, which focuses on temporal interactions between humans and objects localized with bounding boxes in videos. However, the limitation of bounding boxes in detecting non-rigid objects and backgrounds often causes VidSGG systems to miss key details that are crucial for comprehensive video understanding. In contrast, PVSG requires nodes in scene graphs to be grounded by more precise, pixel-level segmentation masks, which facilitate holistic scene understanding. To advance research in this new area, we contribute a high-quality PVSG dataset, which consists of 400 videos (289 third-person + 111 egocentric videos) with totally 150K frames labeled with panoptic segmentation masks as well as fine, temporal scene graphs. We also provide a variety of baseline methods and share useful design practices for future work.

Text-driven generation models are flourishing in video generation and editing. However, face-centric text-to-video generation remains a challenge due to the lack of a suitable dataset containing high-quality videos and highly relevant texts. This paper presents Celeb V- Text, a large-scale, di-verse, and high-quality dataset of facial text-video pairs, to facilitate research on facial text-to- video generation tasks. CelebV-Text comprises 70,000 in-the-wild face video clips with diverse visual content, each paired with 20 texts gen-erated using the proposed semi-automatic text generation strategy. The provided texts are of high quality, describing both static and dynamic attributes precisely. The supe-riority of CelebV- Text over other datasets is demonstrated via comprehensive statistical analysis of the videos, texts, and text-video relevance. The effectiveness and potential of CelebV- Text are further shown through extensive self-evaluation. A benchmark is constructed with representative methods to standardize the evaluation of the facial text-to-video generation task. All data and models are publicly available11Project page: https://celebv-text.github.io.

Text-driven content creation has evolved to be a transformative technique that revolutionizes creativity. Here we study the task of text-driven human video generation, where a video sequence is synthesized from texts describing the appearance and motions of a target performer. Compared to general text-driven video generation, human-centric video generation requires maintaining the appearance of synthesized human while performing complex motions. In this work, we present Text2Performer to generate vivid human videos with articulated motions from texts. Text2Performer has two novel designs: 1) decomposed human representation and 2) diffusion-based motion sampler. First, we decompose the VQVAE latent space into human appearance and pose representation in an unsupervised manner by utilizing the nature of human videos. In this way, the appearance is well maintained along the generated frames. Then, we propose continuous VQ-diffuser to sample a sequence of pose embeddings. Unlike existing VQ-based methods that operate in the discrete space, continuous VQ-diffuser directly outputs the continuous pose embeddings for better motion modeling. Finally, motion-aware masking strategy is designed to mask the pose embeddings spatial-temporally to enhance the temporal coherence. Moreover, to facilitate the task of text-driven human video generation, we contribute a Fashion-Text2Video dataset with manually annotated action labels and text descriptions. Extensive experiments demonstrate that Text2Performer generates high-quality human videos (up to 512 × 256 resolution) with diverse appearances and flexible motions. Our project page is https://yumingj.github.io/projects/Text2Performer.html

Realistic human-centric rendering plays a key role in both computer vision and computer graphics. Rapid progress has been made in the algorithm aspect over the years, yet existing human-centric rendering datasets and benchmarks are rather impoverished in terms of diversity (e.g., outfit's fabric/material, body's interaction with objects, and motion sequences), which are crucial for rendering effect. Researchers are usually constrained to explore and evaluate a small set of rendering problems on current datasets, while real-world applications require methods to be robust across different scenarios. In this work, we present DNA-Rendering, a large-scale, high-fidelity repository of human performance data for neural actor rendering. DNA-Rendering presents several appealing attributes. First, our dataset contains over 1500 human subjects, 5000 motion sequences, and 67.5M frames' data volume. Upon the massive collections, we provide human subjects with grand categories of pose actions, body shapes, clothing, accessories, hairdos, and object intersection, which ranges the geometry and appearance variances from everyday life to professional occasions. Second, we provide rich assets for each subject – 2D/3D human body keypoints, foreground masks, SMPLX models, cloth/accessory materials, multi-view images, and videos. These assets boost the current method's accuracy on downstream rendering tasks. Third, we construct a professional multi-view system to capture data, which contains 60 synchronous cameras with max 4096 × 3000 resolution, 15 fps speed, and stern camera calibration steps, ensuring high-quality resources for task training and evaluation.Along with the dataset, we provide a large-scale and quantitative benchmark in full-scale, with multiple tasks to evaluate the existing progress of novel view synthesis, novel pose animation synthesis, and novel identity rendering methods. In this manuscript, we describe our DNA-Rendering effort as a revealing of new observations, challenges, and future directions to human-centric rendering. The dataset, code, and benchmarks will be publicly available at https://dna-rendering.github.io/.

This paper strives for motion expressions guided video segmentation, which focuses on segmenting objects in video content based on a sentence describing the motion of the objects. Existing referring video object datasets typically focus on salient objects and use language expressions that contain excessive static attributes that could potentially enable the target object to be identified in a single frame. These datasets downplay the importance of motion in video content for language-guided video object segmentation. To investigate the feasibility of using motion expressions to ground and segment objects in videos, we propose a large-scale dataset called MeViS, which contains numerous motion expressions to indicate target objects in complex environments. We benchmarked 5 existing referring video object segmentation (RVOS) methods and conducted a comprehensive comparison on the MeViS dataset. The results show that current RVOS methods cannot effectively address motion expression-guided video segmentation. We further analyze the challenges and propose a baseline approach for the proposed MeViS dataset. The goal of our benchmark is to provide a platform that enables the development of effective language-guided video segmentation algorithms that leverage motion expressions as a primary cue for object segmentation in complex video scenes. The proposed MeViS dataset has been released at https://henghuiding.github.io/MeViS.

Automatic high-quality rendering of anime scenes from complex real-world images is of significant practical value. The challenges of this task lie in the complexity of the scenes, the unique features of anime style, and the lack of high-quality datasets to bridge the domain gap. Despite promising attempts, previous efforts are still incompetent in achieving satisfactory results with consistent semantic preservation, evident stylization, and fine details. In this study, we propose Scenimefy, a novel semi-supervised image-to-image translation framework that addresses these challenges. Our approach guides the learning with structure-consistent pseudo paired data, simplifying the pure unsupervised setting. The pseudo data are derived uniquely from a semantic-constrained StyleGAN leveraging rich model priors like CLIP. We further apply segmentation-guided data selection to obtain high-quality pseudo supervision. A patch-wise contrastive style loss is introduced to improve stylization and fine details. Besides, we contribute a high-resolution anime scene dataset to facilitate future research. Our extensive experiments demonstrate the superiority of our method over state-of-the-art baselines in terms of both perceptual quality and quantitative performance. Project page: https://yuxinn-j.github.io/projects/Scenimefy.html.

Video segmentation aims to segment and track every pixel in diverse scenarios accurately. In this paper, we present Tube-Link, a versatile framework that addresses multiple core tasks of video segmentation with a unified architecture. Our framework is a near-online approach that takes a short subclip as input and outputs the corresponding spatial-temporal tube masks. To enhance the modeling of cross-tube relationships, we propose an effective way to perform tube-level linking via attention along the queries. In addition, we introduce temporal contrastive learning to instance-wise discriminative features for tube-level association. Our approach offers flexibility and efficiency for both short and long video inputs, as the length of each subclip can be varied according to the needs of datasets or scenarios. Tube-Link outperforms existing specialized architectures by a significant margin on five video segmentation datasets. Specifically, it achieves almost 13% relative improvements on VIPSeg and 4% improvements on KITTI-STEP over the strong baseline Video K-Net. When using a ResNet50 backbone on Youtube-VIS2019 and 2021, Tube-Link boosts IDOL by 3% and 4%, respectively. Code is available at https://github.com/lxtGH/Tube-Link.

We propose a novel unsupervised backlit image enhancement method, abbreviated as CLIP-LIT, by exploring the potential of Contrastive Language-Image Pre-Training (CLIP) for pixel-level image enhancement. We show that the openworld CLIP prior not only aids in distinguishing between backlit and well-lit images, but also in perceiving heterogeneous regions with different luminance, facilitating the optimization of the enhancement network. Unlike high-level and image manipulation tasks, directly applying CLIP to enhancement tasks is non-trivial, owing to the difficulty in finding accurate prompts. To solve this issue, we devise a prompt learning framework that first learns an initial prompt pair by constraining the text-image similarity between the prompt (negative/positive sample) and the corresponding image (backlit image/well-lit image) in the CLIP latent space. Then, we train the enhancement network based on the textimage similarity between the enhanced result and the initial prompt pair. To further improve the accuracy of the initial prompt pair, we iteratively fine-tune the prompt learning framework to reduce the distribution gaps between the backlit images, enhanced results, and well-lit images via rank learning, boosting the enhancement performance. Our method alternates between updating the prompt learning framework and enhancement network until visually pleasing results are achieved. Extensive experiments demonstrate that our method outperforms state-of-the-art methods in terms of visual quality and generalization ability, without requiring any paired data.

Mimicking realistic dynamics in 3D garment animations is a challenging task due to the complex nature of multi-layered garments and the variety of outer forces involved. Existing approaches mostly focus on single-layered garments driven by only human bodies and struggle to handle general scenarios. In this paper, we propose a novel data-driven method, called LayersNet, to model garment-level animations as particle-wise interactions in a micro physics system. We improve simulation efficiency by representing garments as patch-level particles in a two-level structural hierarchy. Moreover, we introduce a novel Rotation Equivalent Transformation with Rotation Invariant Attention that leverage the rotation invariance and additivity of physics systems to better model outer forces. To verify the effectiveness of our approach and bridge the gap between experimental environments and real-world scenarios, we introduce a new challenging dataset, D-LAYERS, containing 700K frames of dynamics of 4,900 combinations of multi-layered garments driven by human bodies and randomly sampled wind. Our LayersNet achieves superior performance both quantitatively and qualitatively. Project page: www.mmlab-ntu.com/project/layersnet/index.html.

We aim to address a significant but understudied problem in the anime industry, namely the inbetweening of cartoon line drawings. Inbetweening involves generating intermediate frames between two black-and-white line drawings and is a time-consuming and expensive process that can benefit from automation. However, existing frame interpolation methods that rely on matching and warping whole raster images are unsuitable for line inbetweening and often produce blurring artifacts that damage the intricate line structures. To preserve the precision and detail of the line drawings, we propose a new approach, AnimeInbet, which geometrizes raster line drawings into graphs of endpoints and reframes the inbetweening task as a graph fusion problem with vertex repositioning. Our method can effectively capture the sparsity and unique structure of line drawings while preserving the details during inbetweening. This is made possible via our novel modules, i.e., vertex geometric embedding, a vertex correspondence Transformer, an effective mechanism for vertex repositioning and a visibility predictor. To train our method, we introduce Mixamo-Line240, a new dataset of line drawings with ground truth vectorization and matching labels. Our experiments demonstrate that AnimeInbet synthesizes high-quality, clean, and complete intermediate line drawings, outperforming existing methods quantitatively and qualitatively, especially in cases with large motions. Data and code are available at https://github.com/lisiyao21/AnimeInbet.

Neural Radiance Field (NeRF) significantly degrades when only a limited number of views are available. To complement the lack of 3D information, depth-based models, such as DSNeRF and MonoSDF, explicitly assume the availability of accurate depth maps of multiple views. They linearly scale the accurate depth maps as supervision to guide the predicted depth of few-shot NeRFs. However, accurate depth maps are difficult and expensive to capture due to wide-range depth distances in the wild. This work presents a new Sparse-view NeRF (SparseNeRF) framework that exploits depth priors from real-world inaccurate observations. The inaccurate depth observations are either from pre-trained depth models or coarse depth maps of consumer-level depth sensors. Since coarse depth maps are not strictly scaled to the ground-truth depth maps, we propose a simple yet effective constraint, a local depth ranking method, on NeRFs such that the expected depth ranking of the NeRF is consistent with that of the coarse depth maps in local patches. To preserve the spatial continuity of the estimated depth of NeRF, we further propose a spatial continuity constraint to encourage the consistency of the expected depth continuity of NeRF with coarse depth maps. Surprisingly, with simple depth ranking constraints, SparseNeRF outperforms all state-of-the-art few-shot NeRF methods (including depth-based models) on standard LLFF and DTU datasets. Moreover, we collect a new dataset NVS-RGBD that contains real-world depth maps from Azure Kinect, ZED 2, and iPhone 13 Pro. Extensive experiments on NVS-RGBD dataset also validate the superiority and generaliz-ability of SparseNeRF. Code and dataset are available at https://sparsenerf.github.io/.

Unconditional video generation is a challenging task that involves synthesizing high-quality videos that are both coherent and of extended duration. To address this challenge, researchers have used pretrained StyleGAN image generators for high-quality frame synthesis and focused on motion generator design. The motion generator is trained in an autoregressive manner using heavy 3D convolutional discriminators to ensure motion coherence during video generation. In this paper, we introduce a novel motion generator design that uses a learning-based inversion network for GAN. The encoder in our method captures rich and smooth priors from encoding images to latents, and given the latent of an initially generated frame as guidance, our method can generate smooth future latent by modulating the inversion encoder temporally. Our method enjoys the advantage of sparse training and naturally constrains the generation space of our motion generator with the inversion network guided by the initial frame, eliminating the need for heavy discriminators. Moreover, our method supports style transfer with simple fine-tuning when the encoder is paired with a pretrained StyleGAN generator. Extensive experiments conducted on various benchmarks demonstrate the superiority of our method in generating long and high-resolution videos with decent single-frame quality and temporal consistency. Code is available at https://github.com/johannwyh/StyleInV.

In this work, we focus on open vocabulary instance segmentation to expand a segmentation model to classify and segment instance-level novel categories. Previous approaches have relied on massive caption datasets and complex pipelines to establish one-to-one mappings between image regions and words in captions. However, such methods build noisy supervision by matching non-visible words to image regions, such as adjectives and verbs. Meanwhile, context words are also important for inferring the existence of novel objects as they show high inter-correlations with novel categories. To overcome these limitations, we devise a joint Caption Grounding and Generation (CGG) framework, which incorporates a novel grounding loss that only focuses on matching object nouns to improve learning efficiency. We also introduce a caption generation head that enables additional supervision and contextual modeling as a complementation to the grounding loss. Our analysis and results demonstrate that grounding and generation components complement each other, significantly enhancing the segmentation performance for novel classes. Experiments on the COCO dataset with two settings: Open Vocabulary Instance Segmentation (OVIS) and Open Set Panoptic Segmentation (OSPS) demonstrate the superiority of the CGG. Specifically, CGG achieves a substantial improvement of 6.8% mAP for novel classes without extra data on the OVIS task and 15% PQ improvements for novel classes on the OSPS benchmark.

Recent advances in face manipulation using StyleGAN have produced impressive results. However, StyleGAN is inherently limited to cropped aligned faces at a fixed image resolution it is pre-trained on. In this paper, we propose a simple and effective solution to this limitation by using dilated convolutions to rescale the receptive fields of shallow layers in StyleGAN, without altering any model parameters. This allows fixed-size small features at shallow layers to be extended into larger ones that can accommodate variable resolutions, making them more robust in characterizing unaligned faces. To enable real face inversion and manipulation, we introduce a corresponding encoder that provides the first-layer feature of the extended StyleGAN in addition to the latent style code. We validate the effectiveness of our method using unaligned face inputs of various resolutions in a diverse set of face manipulation tasks, including facial attribute editing, super-resolution, sketch/mask-to-face translation, and face toonification. Project page https://www.mmlab-ntu.com/project/styleganex

In this paper, we address the challenging problem of 3D toonification, which involves transferring the style of an artistic domain onto a target 3D face with stylized geometry and texture. Although fine-tuning a pre-trained 3D GAN on the artistic domain can produce reasonable performance, this strategy has limitations in the 3D domain. In particular, fine-tuning can deteriorate the original GAN latent space, which affects subsequent semantic editing, and requires independent optimization and storage for each new style, limiting flexibility and efficient deployment. To overcome these challenges, we propose DeformToon3d, an effective toonification framework tailored for hierarchical 3D GAN. Our approach decomposes 3D toonification into subproblems of geometry and texture stylization to better preserve the original latent space. Specifically, we devise a novel StyleField that predicts conditional 3D deformation to align a real-space NeRF to the style space for geometry stylization. Thanks to the StyleField formulation, which already handles geometry stylization well, texture stylization can be achieved conveniently via adaptive style mixing that injects information of the artistic domain into the decoder of the pre-trained 3D GAN. Due to the unique design, our method enables flexible style degree control and shape-texture-specific style swap. Furthermore, we achieve efficient training without any real-world 2D-3D training pairs but proxy samples synthesized from off-the-shelf 2D toonification models. Code is released at https://github.com/junzhezhang/DeformToon3D.

Flow-based propagation and spatiotemporal Transformer are two mainstream mechanisms in video inpainting (VI). Despite the effectiveness of these components, they still suffer from some limitations that affect their performance. Previous propagation-based approaches are performed separately either in the image or feature domain. Global image propagation isolated from learning may cause spatial misalignment due to inaccurate optical flow. Moreover, memory or computational constraints limit the temporal range of feature propagation and video Transformer, preventing exploration of correspondence information from distant frames. To address these issues, we propose an improved framework, called ProPainter, which involves enhanced ProPagation and an efficient Transformer. Specifically, we introduce dual-domain propagation that combines the advantages of image and feature warping, exploiting global correspondences reliably. We also propose a mask-guided sparse video Transformer, which achieves high efficiency by discarding unnecessary and redundant tokens. With these components, ProPainter outperforms prior arts by a large margin of 1.46 dB in PSNR while maintaining appealing efficiency.

Expressive human pose and shape estimation (EHPS) unifies body, hands, and face motion capture with numerous applications. Despite encouraging progress, current state-of-the-art methods still depend largely on a confined set of training datasets. In this work, we investigate scaling up EHPS towards the first generalist foundation model (dubbed SMPLer-X), with up to ViT-Huge as the backbone and training with up to 4.5M instances from diverse data sources. With big data and the large model, SMPLer-X exhibits strong performance across diverse test benchmarks and excellent transferability to even unseen environments. 1) For the data scaling, we perform a systematic investigation on 32 EHPS datasets, including a wide range of scenarios that a model trained on any single dataset cannot handle. More importantly, capitalizing on insights obtained from the extensive benchmarking process, we optimize our training scheme and select datasets that lead to a significant leap in EHPS capabilities. 2) For the model scaling, we take advantage of vision transformers to study the scaling law of model sizes in EHPS. Moreover, our finetuning strategy turn SMPLer-X into specialist models, allowing them to achieve further performance boosts. Notably, our foundation model SMPLer-X consistently delivers state-of-the-art results on seven benchmarks such as AGORA (107.2 mm NMVE), UBody (57.4 mm PVE), EgoBody (63.6 mm PVE), and EHF (62.3 mm PVE without finetuning).

With the rise of large-scale models trained on broad data, in-context learning has become a new learning paradigm that has demonstrated significant potential in natural language processing and computer vision tasks. Meanwhile, in-context learning is still largely unexplored in the 3D point cloud domain. Although masked modeling has been successfully applied for in-context learning in 2D vision, directly extending it to 3D point clouds remains a formidable challenge. In the case of point clouds, the tokens themselves are the point cloud positions (coordinates) that are masked during inference. Moreover, position embedding in previous works may inadvertently introduce information leakage. To address these challenges, we introduce a novel framework, named Point-In-Context, designed especially for in-context learning in 3D point clouds, where both inputs and outputs are modeled as coordinates for each task. Additionally, we propose the Joint Sampling module, carefully designed to work in tandem with the general point sampling operator, effectively resolving the aforementioned technical issues. We conduct extensive experiments to validate the versatility and adaptability of our proposed methods in handling a wide range of tasks. Furthermore, with a more effective prompt selection strategy, our framework surpasses the results of individually trained models.

Synthesizing high-fidelity head avatars is a central problem for computer vision and graphics. While head avatar synthesis algorithms have advanced rapidly, the best ones still face great obstacles in real-world scenarios. One of the vital causes is the inadequate datasets -- 1) current public datasets can only support researchers to explore high-fidelity head avatars in one or two task directions; 2) these datasets usually contain digital head assets with limited data volume, and narrow distribution over different attributes, such as expressions, ages, and accessories. In this paper, we present RenderMe-360, a comprehensive 4D human head dataset to drive advance in head avatar algorithms across different scenarios. It contains massive data assets, with 243+ million complete head frames and over 800k video sequences from 500 different identities captured by multi-view cameras at 30 FPS. It is a large-scale digital library for head avatars with three key attributes: 1) High Fidelity: all subjects are captured in 360 degrees via 60 synchronized, high-resolution 2K cameras. 2) High Diversity: The collected subjects vary from different ages, eras, ethnicities, and cultures, providing abundant materials with distinctive styles in appearance and geometry. Moreover, each subject is asked to perform various dynamic motions, such as expressions and head rotations, which further extend the richness of assets. 3) Rich Annotations: the dataset provides annotations with different granularities: cameras' parameters, background matting, scan, 2D/3D facial landmarks, FLAME fitting, and text description. Based on the dataset, we build a comprehensive benchmark for head avatar research, with 16 state-of-the-art methods performed on five main tasks: novel view synthesis, novel expression synthesis, hair rendering, hair editing, and talking head generation. Our experiments uncover the strengths and flaws of state-of-the-art methods. RenderMe-360 opens the door for future exploration in modern head avatars. All of the data, code, and models will be publicly available at https://renderme-360.github.io/.

Exploiting pre-trained diffusion models for restoration has recently become a favored alternative to the traditional task-specific training approach. Previous works have achieved noteworthy success by limiting the solution space using explicit degradation models. However, these methods often fall short when faced with complex degradations as they generally cannot be precisely modeled. In this paper, we introduce $\textit{partial guidance}$, a fresh perspective that is more adaptable to real-world degradations compared to existing works. Rather than specifically defining the degradation process, our approach models the desired properties, such as image structure and color statistics of high-quality images, and applies this guidance during the reverse diffusion process. These properties are readily available and make no assumptions about the degradation process. When combined with a diffusion prior, this partial guidance can deliver appealing results across a range of restoration tasks. Additionally, our method can be extended to handle composite tasks by consolidating multiple high-quality image properties, achieved by integrating the guidance from respective tasks. Experimental results demonstrate that our method not only outperforms existing diffusion-prior-based approaches but also competes favorably with task-specific models.

Diffusion-based image super-resolution (SR) methods are mainly limited by the low inference speed due to the requirements of hundreds or even thousands of sampling steps. Existing acceleration sampling techniques inevitably sacrifice performance to some extent, leading to over-blurry SR results. To address this issue, we propose a novel and efficient diffusion model for SR that significantly reduces the number of diffusion steps, thereby eliminating the need for post-acceleration during inference and its associated performance deterioration. Our method constructs a Markov chain that transfers between the high-resolution image and the low-resolution image by shifting the residual between them, substantially improving the transition efficiency. Additionally, an elaborate noise schedule is developed to flexibly control the shifting speed and the noise strength during the diffusion process. Extensive experiments demonstrate that the proposed method obtains superior or at least comparable performance to current state-of-the-art methods on both synthetic and real-world datasets, \textit{\textbf{even only with 20 sampling steps}}. Our code and model will be made publicly.

Image restoration techniques, spanning from the convolution to the transformer paradigm, have demonstrated robust spatial representation capabilities to deliver high-quality performance.Yet, many of these methods, such as convolution and the Feed Forward Network (FFN) structure of transformers, primarily leverage the basic first-order channel interactions and have not maximized the potential benefits of higher-order modeling. To address this limitation, our research dives into understanding relationships within the channel dimension and introduces a simple yet efficient, high-order channel-wise operator tailored for image restoration. Instead of merely mimicking high-order spatial interaction, our approach offers several added benefits: Efficiency: It adheres to the zero-FLOP and zero-parameter principle, using a spatial-shifting mechanism across channel-wise groups. Simplicity: It turns the favorable channel interaction and aggregation capabilities into element-wise multiplications and convolution units with $1 \times 1$ kernel. Our new formulation expands the first-order channel-wise interactions seen in previous works to arbitrary high orders, generating a hierarchical receptive field akin to a Rubik's cube through the combined action of shifting and interactions. Furthermore, our proposed Rubik's cube convolution is a flexible operator that can be incorporated into existing image restoration networks, serving as a drop-in replacement for the standard convolution unit with fewer parameters overhead. We conducted experiments across various low-level vision tasks, including image denoising, low-light image enhancement, guided image super-resolution, and image de-blurring. The results consistently demonstrate that our Rubik's cube operator enhances performance across all tasks. Code is publicly available at https://github.com/zheng980629/RubikCube.

Large text-to-image diffusion models have exhibited impressive proficiency in generating high-quality images. However, when applying these models to video domain, ensuring temporal consistency across video frames remains a formidable challenge. This paper proposes a novel zero-shot text-guided video-to-video translation framework to adapt image models to videos. The framework includes two parts: key frame translation and full video translation. The first part uses an adapted diffusion model to generate key frames, with hierarchical cross-frame constraints applied to enforce coherence in shapes, textures and colors. The second part propagates the key frames to other frames with temporal-aware patch matching and frame blending. Our framework achieves global style and local texture temporal consistency at a low cost (without re-training or optimization). The adaptation is compatible with existing image diffusion techniques, allowing our framework to take advantage of them, such as customizing a specific subject with LoRA, and introducing extra spatial guidance with ControlNet. Extensive experimental results demonstrate the effectiveness of our proposed framework over existing methods in rendering high-quality and temporally-coherent videos. Code is available at our project page: https://www.mmlab-ntu.com/project/rerender/

We show that pre-trained Generative Adversarial Networks (GANs) such as StyleGAN and BigGAN can be used as a latent bank to improve the performance of image super-resolution. While most existing perceptual-oriented approaches attempt to generate realistic outputs through learning with adversarial loss, our method, Generative LatEnt bANk (GLEAN), goes beyond existing practices by directly leveraging rich and diverse priors encapsulated in a pre-trained GAN. But unlike prevalent GAN inversion methods that require expensive image-specific optimization at runtime, our approach only needs a single forward pass for restoration. GLEAN can be easily incorporated in a simple encoder-bank-decoder architecture with multi-resolution skip connections. Employing priors from different generative models allows GLEAN to be applied to diverse categories (e.g., human faces, cats, buildings, and cars). We further present a lightweight version of GLEAN, named LightGLEAN, which retains only the critical components in GLEAN. Notably, LightGLEAN consists of only 21% of parameters and 35% of FLOPs while achieving comparable image quality. We extend our method to different tasks including image colorization and blind image restoration, and extensive experiments show that our proposed models perform favorably in comparison to existing methods. Codes and models are available at https://github.com/open-mmlab/mmediting.

Reference-based Super-Resolution (Ref-SR) has recently emerged as a promising paradigm to enhance a low-resolution (LR) input image or video by introducing an additional high-resolution (HR) reference image. Existing Ref-SR methods mostly rely on implicit correspondence matching to borrow HR textures from reference images to compensate for the information loss in input images. However, performing local transfer is difficult because of two gaps between input and reference images: the transformation gap (e.g.scale and rotation) and the resolution gap (e.g.HR and LR). To tackle these challenges, we propose$C^{2}$-Matching in this work, which performs explicit robust matching crossing transformation and resolution. 1) To bridge the transformation gap, we propose a contrastive correspondence network, which learns transformation-robust correspondences using augmented views of the input image. 2) To address the resolution gap, we adopt teacher-student correlation distillation, which distills knowledge from the easier HR-HR matching to guide the more ambiguous LR-HR matching. 3) Finally, we design a dynamic aggregation module to address the potential misalignment issue between input images and reference images. In addition, to faithfully evaluate the performance of Reference-based Image Super-Resolution (Ref Image SR) under a realistic setting, we contribute the Webly-Referenced SR (WR-SR) dataset, mimicking the practical usage scenario. We also extend$C^{2}$-Matching to Reference-based Video Super-Resolution (Ref VSR) task, where an image taken in a similar scene serves as the HR reference image. Extensive experiments demonstrate that our proposed$C^{2}$-Matching significantly outperforms state of the arts by up to 0.7dB on the standard CUFED5 benchmark and also boosts the performance of video super-resolution by incorporating the$C^{2}$-Matching component into Video SR pipelines. Notably,$C^{2}$-Matching also shows great generalizability on WR-SR dataset as well as robustness across large scale and rotation transformations. Codes and datasets are available athttps://github.com/yumingj/C2-Matching.

Our proposed music-to-dance framework, Bailando++, addresses the challenges of driving 3D characters to dance in a way that follows the constraints of choreography norms and maintains temporal coherency with different music genres. Bailando++ consists of two components: a choreographic memory that learns to summarize meaningful dancing units from 3D pose sequences, and an actor-critic Generative Pre-trained Transformer (GPT) that composes these units into a fluent dance coherent to the music. In particular, to synchronize the diverse motion tempos and music beats, we introduce an actor-critic-based reinforcement learning scheme to the GPT with a novel beat-align reward function. Additionally, we consider learning human dance poses in the rotation domain to avoid body distortions incompatible with human morphology, and introduce a musical contextual encoding to allow the motion GPT to grasp longer-term patterns of music. Our experiments on the standard benchmark show that Bailando++ achieves state-of-the-art performance both qualitatively and quantitatively, with the added benefit of the unsupervised discovery of human-interpretable dancing-style poses in the choreographic memory.

Recent advances in deep learning have witnessed many successful unsupervised image-to-image translation models that learn correspondences between two visual domains without paired data. However, it is still a great challenge to build robust mappings between various domains especially for those with drastic visual discrepancies. In this paper, we introduce a novel versatile framework, Generative Prior-guided UNsupervised Image-to-image Translation (GP-UNIT), that improves the quality, applicability and controllability of the existing translation models. The key idea of GP-UNIT is to distill the generative prior from pre-trained class-conditional GANs to build coarse-level cross-domain correspondences, and to apply the learned prior to adversarial translations to excavate fine-level correspondences. With the learned multi-level content correspondences, GP-UNIT is able to perform valid translations between both close domains and distant domains. For close domains, GP-UNIT can be conditioned on a parameter to determine the intensity of the content correspondences during translation, allowing users to balance between content and style consistency. For distant domains, semi-supervised learning is explored to guide GP-UNIT to discover accurate semantic correspondences that are hard to learn solely from the appearance. We validate the superiority of GP-UNIT over state-of-the-art translation models in robust, high-quality and diversified translations between various domains through extensive experiments.

Generalization to out-of-distribution (OOD) data is a capability natural to humans yet challenging for machines to reproduce. This is because most learning algorithms strongly rely on the i.i.d. assumption on source/target data, which is often violated in practice due to domain shift. Domain generalization (DG) aims to achieve OOD generalization by using only source data for model learning. Over the last ten years, research in DG has made great progress, leading to a broad spectrum of methodologies, e.g., those based on domain alignment, meta-learning, data augmentation, or ensemble learning, to name a few; DG has also been studied in various application areas including computer vision, speech recognition, natural language processing, medical imaging, and reinforcement learning. In this paper, for the first time a comprehensive literature review in DG is provided to summarize the developments over the past decade. Specifically, we first cover the background by formally defining DG and relating it to other relevant fields like domain adaptation and transfer learning. Then, we conduct a thorough review into existing methods and theories. Finally, we conclude this survey with insights and discussions on future research directions.

Existing automated dubbing methods are usually designed for Professionally Generated Content (PGC) production, which requires massive training data and training time to learn a person-specific audio-video mapping. In this paper, we investigate an audio-driven dubbing method that is more feasible for User Generated Content (UGC) production. There are two unique challenges to design a method for UGC: 1) the appearances of speakers are diverse and arbitrary as the method needs to generalize across users; 2) the available video data of one speaker are very limited. In order to tackle the above challenges, we first introduce a new Style Translation Network to integrate the speaking style of the target and the speaking content of the source via a cross-modal AdaIN module. It enables our model to quickly adapt to a new speaker. Then, we further develop a semi-parametric video renderer, which takes full advantage of the limited training data of the unseen speaker via a video-level retrieve-warp-refine pipeline. Finally, we propose a temporal regularization for the semi-parametric renderer, generating more continuous videos. Extensive experiments show that our method generates videos that accurately preserve various speaking styles, yet with considerably lower amount of training data and training time in comparison to existing methods. Besides, our method achieves a faster testing speed than most recent methods.

We present a novel framework that brings the 3D motion retargeting task from controlled environments to in-the-wild scenarios. In particular, our method is capable of retargeting body motion from a character in a 2D monocular video to a 3D character without using any motion capture system or 3D reconstruction procedure. It is designed to leverage massive online videos for unsupervised training, needless of 3D annotations or motion-body pairing information. The proposed method is built upon two novel canonicalization operations, structure canonicalization and view canonicalization. Trained with the canonicalization operations and the derived regularizations, our method learns to factorize a skeleton sequence into three independent semantic subspaces, i.e., motion, structure, and view angle. The disentangled representation enables motion retargeting from 2D to 3D with high precision. Our method achieves superior performance on motion transfer benchmarks with large body variations and challenging actions. Notably, the canonicalized skeleton sequence could serve as a disentangled and interpretable representation of human motion that benefits action analysis and motion retrieval.

The diversity and complexity of degradations in real-world video super-resolution (VSR) pose non-trivial challenges in inference and training. First, while long-term propagation leads to improved performance in cases of mild degradations, severe in-the-wild degradations could be exaggerated through propagation, impairing output quality. To balance the tradeoff between detail synthesis and artifact suppression, we found an image precleaning stage in-dispensable to reduce noises and artifacts prior to propagation. Equipped with a carefully designed cleaning module, our RealBasicVSR outperforms existing methods in both quality and efficiency (Fig. 1). Second, real-world VSR models are often trained with diverse degradations to improve generalizability, requiring increased batch size to produce a stable gradient. Inevitably, the increased computational burden results in various problems, including 1) speed-performance tradeoff and 2) batch-length trade-off. To alleviate the first tradeoff, we propose a stochastic degradation scheme that reduces up to 40% of training time without sacrificing performance. We then analyze different training settings and suggest that employing longer sequences rather than larger batches during training allows more effective uses of temporal information, leading to more stable performance during inference. To facilitate fair comparisons, we propose the new VideoLQ dataset, which contains a large variety of real-world low-quality video sequences containing rich textures and patterns. Our dataset can serve as a common ground for benchmarking. Code, models, and the dataset are publicly available at https://github.com/ckkelvinchan/RealBasicVSR.

A recurrent structure is a popular framework choice for the task of video super-resolution. The state-of-the-art method BasicVSR adopts bidirectional propagation with feature alignment to effectively exploit information from the entire input video. In this study, we redesign BasicVsr by proposing second-order grid propagation and flow-guided deformable alignment. We show that by empowering the re-current framework with enhanced propagation and align-ment, one can exploit spatiotemporal information across misaligned video frames more effectively. The new components lead to an improved performance under a simi-lar computational constraint. In particular, our model Ba-sicVSR++ surpasses BasicVSR by a significant 0.82 dB in PSNR with similar number of parameters. BasicVSR++ is generalizable to other video restoration tasks, and obtains three champions and one first runner-up in NTIRE 2021 video restoration challenge.

This article addresses the problem of distilling knowledge from a large teacher model to a slim student network for LiDAR semantic segmentation. Directly employing previous distillation approaches yields inferior results due to the intrinsic challenges of point cloud, i.e., sparsity, randomness and varying density. To tackle the aforementioned problems, we propose the Point-to-Voxel Knowledge Distillation (PVD), which transfers the hidden knowledge from both point level and voxel level. Specifically, we first leverage both the pointwise and voxelwise output distillation to complement the sparse supervision signals. Then, to better exploit the structural information, we divide the whole point cloud into several supervoxels and design a difficulty-aware sampling strategy to more frequently sample supervoxels containing less-frequent classes and faraway objects. On these supervoxels, we propose inter-point and inter-voxel affinity distillation, where the similarity information between points and voxels can help the student model better capture the structural information of the surrounding environment. We conduct extensive experiments on two popular LiDAR segmentation benchmarks, i.e., nuScenes [3] and SemanticKITTI [1]. On both benchmarks, our PVD-consistently outperforms previous distillation approaches by a large margin on three representative backbones, i.e., Cylinder3D [36], [37], SPVNAS [25] and MinkowskiNet [5]. Notably, on the challenging nuScenes and SemanticKITTI datasets, our method can achieve roughly 75% MACs reduction and 2× speedup on the competitive Cylinder3D model and rank 1st on the SemanticKITTI leaderboard among all published algorithms11https://competitions.codalab.org/competitions/20331#results (single-scan competition) till 2021-11-18 04:00 Pacific Time, and our method is termed Point-Voxel-KD. Our method (PV-KD) ranks 3rd on the multi-scan challenge till 2021-12-1 00:00 Pacific Time.. Our code is available at https://github.com/cardwing/Codes-for-PVKD.

This paper presents Video K-Net, a simple, strong, and unified framework for fully end-to-end video panoptic seg-mentation. The method is built upon K-Net, a method that unifies image segmentation via a group of learnable ker-nels. We observe that these learnable kernels from K-Net, which encode object appearances and contexts, can naturally associate identical instances across video frames. Motivated by this observation, Video K-Net learns to simultaneously segment and track “things” and “stuff” in a video with simple kernel-based appearance modeling and cross-temporal kernel interaction. Despite the simplicity, it achieves state-of-the-art video panoptic segmentation results on Citscapes-VPS and KITTI-STEP without bells and whistles. In particular on KITTI-STEP, the simple method can boost almost 12% relative improvements over previous methods. We also validate its generalization on video semantic segmentation, where we boost various baselines by 2% on the VSPW dataset. Moreover, we extend K-Net into clip-level video framework for video instance segmentation where we obtain 40.5% for ResNet50 backbone and 51.5% mAP for Swin-base on YouTube-2019 validation set. We hope this simple yet effective method can serve as a new flexible baseline in video segmentation.11Both code and models are released at here.

Driving 3D characters to dance following a piece of music is highly challenging due to the spatial constraints applied to poses by choreography norms. In addition, the generated dance sequence also needs to maintain temporal coherency with different music genres. To tackle these challenges, we propose a novel music-to-dance framework, Bailando, with two powerful components: 1) a choreographic memory that learns to summarize meaningful dancing units from 3D pose sequence to a quantized codebook, 2) an actor-critic Generative Pre-trained Transformer (GPT) that composes these units to a fluent dance coherent to the music. With the learned choreographic memory, dance generation is realized on the quantized units that meet high choreography standards, such that the generated dancing sequences are confined within the spatial constraints. To achieve synchronized alignment between diverse motion tempos and music beats, we introduce an actor-critic-based reinforcement learning scheme to the GPT with a newly-designed beat-align reward function. Extensive experiments on the standard benchmark demonstrate that our proposed framework achieves state-of-the-art performance both qualitatively and quantitatively. Notably, the learned choreographic memory is shown to discover human-interpretable dancing-style poses in an unsupervised manner. Code and video demo are available at https://github.com/lisiyao21/Bailando/

Recent advances like StyleGAN have promoted the growth of controllable facial editing. To address its core challenge of attribute decoupling in a single latent space, attempts have been made to adopt dual-space GAN for better disentanglement of style and content representations. Nonetheless, these methods are still incompetent to obtain plausible editing results with high controllability, especially for complicated attributes. In this study, we highlight the importance of interaction in a dual-space GAN for more controllable editing. We propose TransEditor, a novel Transformer-based framework to enhance such interaction. Besides, we develop a new dual-space editing and inversion strategy to provide additional editing flexibility. Extensive experiments demonstrate the superiority of the proposed framework in image quality and editing capability, suggesting the effectiveness of TransEditor for highly controllable facial editing. Code and models are publicly available at https://github.com/BillyXYB/TransEditor.

Recent studies on StyleGAN show high performance on artistic portrait generation by transfer learning with limited data. In this paper, we explore more challenging exemplar-based high-resolution portrait style transfer by introducing a novel DualStyleGAN with flexible control of dual styles of the original face domain and the extended artistic portrait domain. Different from StyleGAN, DualStyleGAN provides a natural way of style transfer by characterizing the content and style of a portrait with an intrinsic style path and a new extrinsic style path, respectively. The del-icately designed extrinsic style path enables our model to modulate both the color and complex structural styles hierarchically to precisely pastiche the style example. Furthermore, a novel progressive fine-tuning scheme is introduced to smoothly transform the generative space of the model to the target domain, even with the above modifications on the network architecture. Experiments demonstrate the superiority of DualStyleGAN over state-of-the-art methods in high-quality portrait style transfer and flexible stylecontrol. Code is available at https://github.com/williamyang1991/DualStyleGAN.

Unsupervised image-to-image translation aims to learn the translation between two visual domains without paired data. Despite the recent progress in image translation models, it remains challenging to build mappings between complex domains with drastic visual discrepancies. In this work, we present a novel framework, Generative Priorguided UNsupervised Image-to-image Translation (GP-UNIT), to improve the overall quality and applicability of the translation algorithm. Our key insight is to leverage the generative prior from pre-trained class-conditional GANs (e.g., BigGAN) to learn rich content correspondences across various domains. We propose a novel coarse-to-fine scheme: we first distill the generative prior to capture a robust coarse-level content representation that can link objects at an abstract semantic level, based on which finelevel content features are adaptively learned for more accurate multi-level content correspondences. Extensive experiments demonstrate the superiority of our versatile framework over state-of-the-art methods in robust, high-quality and diversified translations, even for challenging and distant domains. Code is available at https://github.com/williamyang1991/GP-UNIT.

With the rise of powerful pre-trained vision-language models like CLIP, it becomes essential to investigate ways to adapt these models to downstream datasets. A recently proposed method named Context Optimization (CoOp) introduces the concept of prompt learning—a recent trend in NLP—to the vision domain for adapting pre-trained vision-language models. Specifically, CoOp turns context words in a prompt into a set of learnable vectors and, with only a few labeled images for learning, can achieve huge improvements over intensively-tuned manual prompts. In our study we identify a critical problem of CoOp: the learned context is not generalizable to wider unseen classes within the same dataset, suggesting that CoOp overfits base classes observed during training. To address the problem, we propose Conditional Context Optimization (CoCoOp), which extends CoOp by further learning a lightweight neural network to generate for each image an input-conditional token (vector). Compared to CoOp's static prompts, our dynamic prompts adapt to each instance and are thus less sensitive to class shift. Extensive experiments show that CoCoOp generalizes much better than CoOp to unseen classes, even showing promising transferability beyond a single dataset; and yields stronger domain generalization performance as well. Code is available at https://github.com/KaiyangZhou/CoOp.

Artificial lights commonly leave strong lens flare artifacts on images captured at night. Nighttime flare not only affects the visual quality but also degrades the performance of vision algorithms. Existing flare removal methods mainly focus on removing daytime flares and fail in nighttime. Nighttime flare removal is challenging because of the unique luminance and spectrum of artificial lights and the diverse patterns and image degradation of the flares captured at night. The scarcity of nighttime flare removal datasets limits the research on this crucial task. In this paper, we introduce, Flare7K, the first nighttime flare removal dataset, which is generated based on the observation and statistics of real-world nighttime lens flares. It offers 5,000 scattering and 2,000 reflective flare images, consisting of 25 types of scattering flares and 10 types of reflective flares. The 7,000 flare patterns can be randomly added to flare-free images, forming the flare-corrupted and flare-free image pairs. With the paired data, we can train deep models to restore flare-corrupted images taken in the real world effectively. Apart from abundant flare patterns, we also provide rich annotations, including the labeling of light source, glare with shimmer, reflective flare, and streak, which are commonly absent from existing datasets. Hence, our dataset can facilitate new work in nighttime flare removal and more fine-grained analysis of flare patterns. Extensive experiments show that our dataset adds diversity to existing flare datasets and pushes the frontier of nighttime flare removal.

Visual correspondence of 2D animation is the core of many applications and deserves careful study. Existing correspondence datasets for 2D cartoon suffer from simple frame composition and monotonic movements, making them insufficient to simulate real animations. In this work, we present a new 2D animation visual correspondence dataset, AnimeRun, by converting open source 3D movies to full scenes in 2D style, including simultaneous moving background and interactions of multiple subjects. Statistics show that our proposed dataset not only resembles real anime more in image composition, but also possesses richer and more complex motion patterns compared to existing datasets. With this dataset, we establish a comprehensive benchmark by evaluating several existing optical flow and segment matching methods, and analyze shortcomings of these methods on animation data. Data are available at https://lisiyao21.github.io/projects/AnimeRun.

Blind face restoration is a highly ill-posed problem that often requires auxiliary guidance to 1) improve the mapping from degraded inputs to desired outputs, or 2) complement high-quality details lost in the inputs. In this paper, we demonstrate that a learned discrete codebook prior in a small proxy space largely reduces the uncertainty and ambiguity of restoration mapping by casting \textit{blind face restoration} as a \textit{code prediction} task, while providing rich visual atoms for generating high-quality faces. Under this paradigm, we propose a Transformer-based prediction network, named \textit{CodeFormer}, to model the global composition and context of the low-quality faces for code prediction, enabling the discovery of natural faces that closely approximate the target faces even when the inputs are severely degraded. To enhance the adaptiveness for different degradation, we also propose a controllable feature transformation module that allows a flexible trade-off between fidelity and quality. Thanks to the expressive codebook prior and global modeling, \textit{CodeFormer} outperforms the state of the arts in both quality and fidelity, showing superior robustness to degradation. Extensive experimental results on synthetic and real-world datasets verify the effectiveness of our method.

Existing convolutional neural networks widely adopt spatial down-/up-sampling for multi-scale modeling. However, spatial up-sampling operators (e.g., interpolation, transposed convolution, and un-pooling) heavily depend on local pixel attention, incapably exploring the global dependency. In contrast, the Fourier domain is in accordance with the nature of global modeling according to the spectral convolution theorem. Unlike the spatial domain that easily performs up-sampling with the property of local similarity, up-sampling in the Fourier domain is more challenging as it does not follow such a local property. In this study, we propose a theoretically feasible Deep Fourier Up-Sampling (FourierUp) to solve these issues. We revisit the relationships between spatial and Fourier domains and reveal the transform rules on the features of different resolutions in the Fourier domain, which provide key insights for FourierUp's designs. FourierUp as a generic operator consists of three key components: 2D discrete Fourier transform, Fourier dimension increase rules, and 2D inverse Fourier transform, which can be directly integrated with existing networks. Extensive experiments across multiple computer vision tasks, including object detection, image segmentation, image de-raining, image dehazing, and guided image super-resolution, demonstrate the consistent performance gains obtained by introducing our FourierUp. Code will be publicly available.

Low-light image enhancement (LLIE) aims at improving the perception or interpretability of an image captured in an environment with poor illumination. Recent advances in this area are dominated by deep learning-based solutions, where many learning strategies, network structures, loss functions, training data, etc. have been employed. In this paper, we provide a comprehensive survey to cover various aspects ranging from algorithm taxonomy to unsolved open issues. To examine the generalization of existing methods, we propose a low-light image and video dataset, in which the images and videos are taken by different mobile phones' cameras under diverse illumination conditions. Besides, for the first time, we provide a unified online platform that covers many popular LLIE methods, of which the results can be produced through a user-friendly web interface. In addition to qualitative and quantitative evaluation of existing methods on publicly available and our proposed datasets, we also validate their performance in face detection in the dark. This survey together with the proposed dataset and online platform could serve as a reference source for future study and promote the development of this research field. The proposed platform and dataset as well as the collected methods, datasets, and evaluation metrics are publicly available and will be regularly updated. Project page: https://www.mmlab-ntu.com/project/lliv_survey/index.html.

This paper presents a novel method, Zero-Reference Deep Curve Estimation (Zero-DCE), which formulates light enhancement as a task of image-specific curve estimation with a deep network. Our method trains a lightweight deep network, DCE-Net, to estimate pixel-wise and high-order curves for dynamic range adjustment of a given image. The curve estimation is specially designed, considering pixel value range, monotonicity, and differentiability. Zero-DCE is appealing in its relaxed assumption on reference images, i.e., it does not require any paired or even unpaired data during training. This is achieved through a set of carefully formulated non-reference loss functions, which implicitly measure the enhancement quality and drive the learning of the network. Despite its simplicity, we show that it generalizes well to diverse lighting conditions. Our method is efficient as image enhancement can be achieved by an intuitive and simple nonlinear curve mapping. We further present an accelerated and light version of Zero-DCE, called Zero-DCE++, that takes advantage of a tiny network with just 10K parameters. Zero-DCE++ has a fast inference speed (1000/11 FPS on a single GPU/CPU for an image of size 1200×900×3) while keeping the enhancement performance of Zero-DCE. Extensive experiments on various benchmarks demonstrate the advantages of our method over state-of-the-art methods qualitatively and quantitatively. Furthermore, the potential benefits of our method to face detection in the dark are discussed. The source code is made publicly available at https://li-chongyi.github.io/Proj_Zero-DCE++.html.

Learning a good image prior is a long-term goal for image restoration and manipulation. While existing methods like deep image prior (DIP) capture low-level image statistics, there are still gaps toward an image prior that captures rich image semantics including color, spatial coherence, textures, and high-level concepts. This work presents an effective way to exploit the image prior captured by a generative adversarial network (GAN) trained on large-scale natural images. As shown in Fig. 1, the deep generative prior (DGP) provides compelling results to restore missing semantics, e.g., color, patch, resolution, of various degraded images. It also enables diverse image manipulation including random jittering, image morphing, and category transfer. Such highly flexible restoration and manipulation are made possible through relaxing the assumption of existing GAN inversion methods, which tend to fix the generator. Notably, we allow the generator to be fine-tuned on-the-fly in a progressive manner regularized by feature distance obtained by the discriminator in GAN. We show that these easy-to-implement and practical changes help preserve the reconstruction to remain in the manifold of nature images, and thus lead to more precise and faithful reconstruction for real images. Code is available at https://github.com/XingangPan/deep-generative-prior.

Feature reassembly, i.e. feature downsampling and upsampling, is a key operation in a number of modern convolutional network architectures, e.g., residual networks and feature pyramids. Its design is critical for dense prediction tasks such as object detection and semantic/instance segmentation. In this work, we propose unified Content-Aware ReAssembly of FEatures (CARAFE++), a universal, lightweight, and highly effective operator to fulfill this goal. CARAFE++ has several appealing properties: (1) Unlike conventional methods such as pooling and interpolation that only exploit sub-pixel neighborhood, CARAFE++ aggregates contextual information within a large receptive field. (2) Instead of using a fixed kernel for all samples (e.g. convolution and deconvolution), CARAFE++ generates adaptive kernels on-the-fly to enable instance-specific content-aware handling. (3) CARAFE++ introduces little computational overhead and can be readily integrated into modern network architectures. We conduct comprehensive evaluations on standard benchmarks in object detection, instance/semantic segmentation, and image inpainting. CARAFE++ shows consistent and substantial gains on mainstream methods across all the tasks with negligible computational overhead. It shows great potential to serve as a strong building block for modern deep networks.

Very deep Convolutional Neural Networks (CNNs) have greatly improved the performance on various image restoration tasks. However, this comes at a price of increasing computational burden, hence limiting their practical usages. We observe that some corrupted image regions are inherently easier to restore than others since the distortion and content vary within an image. To leverage this, we propose Path-Restore, a multi-path CNN with a pathfinder that can dynamically select an appropriate route for each image region. We train the pathfinder using reinforcement learning with a difficulty-regulated reward. This reward is related to the performance, complexity and "the difficulty of restoring a region". A policy mask is further investigated to jointly process all the image regions. We conduct experiments on denoising and mixed restoration tasks. The results show that our method achieves comparable or superior performance to existing approaches with less computational cost. In particular, Path-Restore is effective for real-world denoising, where the noise distribution varies across different regions on a single image. Compared to the state-of-the-art RIDNet [1], our method achieves comparable performance and runs 2.7x faster on the realistic Darmstadt Noise Dataset [2]. Models and codes are available on the project page: https://www.mmlab-ntu.com/project/pathrestore/.

We present a method to edit a target portrait footage by taking a sequence of audio as input to synthesize a photo-realistic video. This method is unique because it is highly dynamic. It does not assume a person-specific rendering network yet capable of translating one source audio into one random chosen video output within a set of speech videos. Instead of learning a highly heterogeneous and nonlinear mapping from audio to the video directly, we first factorize each target video frame into orthogonal parameter spaces,i.e., expression, geometry, and pose, via monocular 3D face reconstruction. Next, a recurrent network is introduced to translate source audio into expression parameters that are primarily related to the audio content. The audio-translated expression parameters are then used to synthesize a photo-realistic human subject in each video frame, with the movement of the mouth regions precisely mapped to the source audio. The geometry and pose parameters of the target human portrait are retained, therefore preserving the context of the original video footage. Finally, we introduce a novel video rendering network and a dynamic programming method to construct a temporally coherent and photo-realistic video. Extensive experiments demonstrate the superiority of our method over existing approaches. Our method is end-to-end learnable and robust to voice variations in the source audio.

Deep neural networks have achieved remarkable progress in single-image 3D human reconstruction. However, existing methods still fall short in predicting rare poses. The reason is that most of the current models perform regression based on a single human prototype, which is similar to common poses while far from the rare poses. In this work, we 1) identify and analyze this learning obstacle and 2) propose a prototype memory-augmented network, PM-Net, that effectively improves performances of predicting rare poses. The core of our framework is a memory module that learns and stores a set of 3D human prototypes capturing local distributions for either common poses or rare poses. With this formulation, the regression starts from a better initialization, which is relatively easier to converge. Extensive experiments on several widely employed datasets demonstrate the proposed framework's effectiveness compared to other state-of-the-art methods. Notably, our approach significantly improves the models' performances on rare poses while generating comparable results on other samples.

Generating high-quality and diverse human images is an important yet challenging task in vision and graphics. However, existing generative models often fall short under the high diversity of clothing shapes and textures. Furthermore, the generation process is even desired to be intuitively controllable for layman users. In this work, we present a text-driven controllable framework, Text2Human, for a high-quality and diverse human generation. We synthesize full-body human images starting from a given human pose with two dedicated steps. 1) With some texts describing the shapes of clothes, the given human pose is first translated to a human parsing map. 2) The final human image is then generated by providing the system with more attributes about the textures of clothes. Specifically, to model the diversity of clothing textures, we build a hierarchical texture-aware codebook that stores multi-scale neural representations for each type of texture. The codebook at the coarse level includes the structural representations of textures, while the codebook at the fine level focuses on the details of textures. To make use of the learned hierarchical codebook to synthesize desired images, a diffusion-based transformer sampler with mixture of experts is firstly employed to sample indices from the coarsest level of the codebook, which then is used to predict the indices of the codebook at finer levels. The predicted indices at different levels are translated to human images by the decoder learned accompanied with hierarchical codebooks. The use of mixture-of-experts allows for the generated image conditioned on the fine-grained text input. The prediction for finer level indices refines the quality of clothing textures. Extensive quantitative and qualitative evaluations demonstrate that our proposed Text2Human framework can generate more diverse and realistic human images compared to state-of-the-art methods. Our project page is https://yumingj.github.io/projects/Text2Human.html. Code and pretrained models are available at https://github.com/yumingj/Text2Human.

Generating high-quality artistic portrait videos is an important and desirable task in computer graphics and vision. Although a series of successful portrait image toonification models built upon the powerful StyleGAN have been proposed, these image-oriented methods have obvious limitations when applied to videos, such as the fixed frame size, the requirement of face alignment, missing non-facial details and temporal inconsistency. In this work, we investigate the challenging controllable high-resolution portrait video style transfer by introducing a novel VToonify framework. Specifically, VToonify leverages the mid- and high-resolution layers of StyleGAN to render high-quality artistic portraits based on the multi-scale content features extracted by an encoder to better preserve the frame details. The resulting fully convolutional architecture accepts non-aligned faces in videos of variable size as input, contributing to complete face regions with natural motions in the output. Our framework is compatible with existing StyleGAN-based image toonification models to extend them to video toonification, and inherits appealing features of these models for flexible style control on color and intensity. This work presents two instantiations of VToonify built upon Toonify and DualStyleGAN for collection-based and exemplar-based portrait video style transfer, respectively. Extensive experimental results demonstrate the effectiveness of our proposed VToonify framework over existing methods in generating high-quality and temporally-coherent artistic portrait videos with flexible style controls. Code and pretrained models are available at our project page: www.mmlab-ntu.com/project/vtoonify/.

3D interacting hand reconstruction is essential to facilitate human-machine interaction and human behaviors understanding. Previous works in this field either rely on auxiliary inputs such as depth images or they can only handle a single hand if monocular single RGB images are used. Single-hand methods tend to generate collided hand meshes, when applied to closely interacting hands, since they cannot model the interactions between two hands explicitly. In this paper, we make the first attempt to reconstruct 3D interacting hands from monocular single RGB images. Our method can generate 3D hand meshes with both precise 3D poses and minimal collisions. This is made possible via a two-stage framework. Specifically, the first stage adopts a convolutional neural network to generate coarse predictions that tolerate collisions but encourage pose-accurate hand meshes. The second stage progressively ameliorates the collisions through a series of factorized refinements while retaining the preciseness of 3D poses. We carefully investigate potential implementations for the factorized refinement, considering the trade-off between efficiency and accuracy. Extensive quantitative and qualitative results on large-scale datasets such as InterHand2.6M demonstrate the effectiveness of the proposed approach.

Deformable convolution, originally proposed for the adaptation to geometric variations of objects, has recently shown compelling performance in aligning multiple frames and is increasingly adopted for video super-resolution. Despite its remarkable performance, its underlying mechanism for alignment remains unclear. In this study, we carefully investigate the relation between deformable alignment and the classic flow-based alignment. We show that deformable convolution can be decomposed into a combination of spatial warping and convolution. This decomposition reveals the commonality of deformable alignment and flow-based alignment in formulation, but with a key difference in their offset diversity. We further demonstrate through experiments that the increased diversity in deformable alignment yields better-aligned features, and hence significantly improves the quality of video super-resolution output. Based on our observations, we propose an offset-fidelity loss that guides the offset learning with optical flow. Experiments show that our loss successfully avoids the overflow of offsets and alleviates the instability problem of deformable alignment. Aside from the contributions to deformable alignment, our formulation inspires a more flexible approach to introduce offset diversity to flow-based alignment, improving its performance.

Reference-based Super-Resolution (Ref-SR) has recently emerged as a promising paradigm to enhance a low-resolution (LR) input image by introducing an additional high-resolution (HR) reference image. Existing Ref-SR methods mostly rely on implicit correspondence matching to borrow HR textures from reference images to compensate for the information loss in input images. However, performing local transfer is difficult because of two gaps between input and reference images: the transformation gap (e.g. scale and rotation) and the resolution gap (e.g. HR and LR). To tackle these challenges, we propose C2-Matching in this work, which produces explicit robust matching crossing transformation and resolution. 1) For the transformation gap, we propose a contrastive correspondence network, which learns transformation-robust correspondences using augmented views of the input image. 2) For the resolution gap, we adopt a teacher-student correlation distillation, which distills knowledge from the easier HR-HR matching to guide the more ambiguous LR-HR matching. 3) Finally, we design a dynamic aggregation module to address the potential misalignment issue. In addition, to faithfully evaluate the performance of Ref-SR under a realistic setting, we contribute the Webly-Referenced SR (WR-SR) dataset, mimicking the practical usage scenario. Extensive experiments demonstrate that our proposed C2-Matching significantly outperforms state of the arts by over 1dB on the standard CUFED5 benchmark. Notably, it also shows great generalizability on WR-SR dataset as well as robustness across large scale and rotation transformations1.

We show that pre-trained Generative Adversarial Networks (GANs), e.g., StyleGAN, can be used as a latent bank to improve the restoration quality of large-factor image super-resolution (SR). While most existing SR approaches attempt to generate realistic textures through learning with adversarial loss, our method, Generative LatEnt bANk (GLEAN), goes beyond existing practices by directly leveraging rich and diverse priors encapsulated in a pre-trained GAN. But unlike prevalent GAN inversion methods that require expensive image-specific optimization at runtime, our approach only needs a single forward pass to generate the upscaled image. GLEAN can be easily incorporated in a simple encoder-bank-decoder architecture with multi-resolution skip connections. Switching the bank allows the method to deal with images from diverse categories, e.g., cat, building, human face, and car. Images upscaled by GLEAN show clear improvements in terms of fidelity and texture faithfulness in comparison to existing methods as shown in Fig. 1.

Video super-resolution (VSR) approaches tend to have more components than the image counterparts as they need to exploit the additional temporal dimension. Complex designs are not uncommon. In this study, we wish to untangle the knots and reconsider some most essential components for VSR guided by four basic functionalities, i.e., Propagation, Alignment, Aggregation, and Upsampling. By reusing some existing components added with minimal redesigns, we show a succinct pipeline, BasicVSR, that achieves appealing improvements in terms of speed and restoration quality in comparison to many state-of-the-art algorithms. We conduct systematic analysis to explain how such gain can be obtained and discuss the pitfalls. We further show the extensibility of BasicVSR by presenting an information-refill mechanism and a coupled propagation scheme to facilitate information aggregation. The BasicVSR and its extension, IconVSR, can serve as strong baselines for future VSR approaches.

Development of Under-Display Camera (UDC) systems provides a true bezel-less and notch-free viewing experience on smartphones (and TV, laptops, tablets), while allowing images to be captured from the selfie camera embedded underneath. In a typical UDC system, the microstructure of the semi-transparent organic light-emitting diode (OLED) pixel array attenuates and diffracts the incident light on the camera, resulting in significant image quality degradation. Oftentimes, noise, flare, haze, and blur can be observed in UDC images. In this work, we aim to analyze and tackle the aforementioned degradation problems. We define a physics-based image formation model to better understand the degradation. In addition, we utilize one of the world’s first commodity UDC smartphone prototypes to measure the real-world Point Spread Function (PSF) of the UDC system, and provide a model-based data synthesis pipeline to generate realistically degraded images. We specially design a new domain knowledge-enabled Dynamic Skip Connection Network (DISCNet) to restore the UDC images. We demonstrate the effectiveness of our method through extensive experiments on both synthetic and real UDC data. Our physics-based image formation model and proposed DISCNet can provide foundations for further exploration in UDC image restoration, and even for general diffraction artifact removal in a broader sense.1

Despite previous success in generating audio-driven talking heads, most of the previous studies focus on the correlation between speech content and the mouth shape. Facial emotion, which is one of the most important features on natural human faces, is always neglected in their methods. In this work, we present Emotional Video Portraits (EVP), a system for synthesizing high-quality video portraits with vivid emotional dynamics driven by audios. Specifically, we propose the Cross-Reconstructed Emotion Disentanglement technique to decompose speech into two decoupled spaces, i.e., a duration-independent emotion space and a duration- dependent content space. With the disentangled features, dynamic 2D emotional facial landmarks can be deduced. Then we propose the Target-Adaptive Face Synthesis technique to generate the final high-quality video portraits, by bridging the gap between the deduced landmarks and the natural head poses of target videos. Extensive experiments demonstrate the effectiveness of our method both qualitatively and quantitatively.1

In the animation industry, cartoon videos are usually produced at low frame rate since hand drawing of such frames is costly and time-consuming. Therefore, it is desirable to develop computational models that can automatically interpolate the in-between animation frames. However, existing video interpolation methods fail to produce satisfying results on animation data. Compared to natural videos, animation videos possess two unique characteristics that make frame interpolation difficult: 1) cartoons comprise lines and smooth color pieces. The smooth areas lack textures and make it difficult to estimate accurate motions on animation videos. 2) cartoons express stories via exaggeration. Some of the motions are non-linear and extremely large. In this work, we formally define and study the animation video interpolation problem for the first time. To address the aforementioned challenges, we propose an effective framework, AnimeInterp, with two dedicated modules in a coarse-to-fine manner. Specifically, 1) Segment-Guided Matching resolves the "lack of textures" challenge by exploiting global matching among color pieces that are piece-wise coherent. 2) Recurrent Flow Refinement resolves the "non-linear and extremely large motion" challenge by recur-rent predictions using a transformer-like architecture. To facilitate comprehensive training and evaluations, we build a large-scale animation triplet dataset, ATD-12K, which comprises 12,000 triplets with rich annotations. Extensive experiments demonstrate that our approach outperforms existing state-of-the-art interpolation methods for animation videos. Notably, AnimeInterp shows favorable perceptual quality and robustness for animation scenarios in the wild. The proposed dataset and code are available at https://github.com/lisiyao21/AnimeInterp/.

We present a new application direction named Pareidolia Face Reenactment, which is defined as animating a static illusory face to move in tandem with a human face in the video. For the large differences between pareidolia face reenactment and traditional human face reenactment, two main challenges are introduced, i.e., shape variance and texture variance. In this work, we propose a novel Parametric Unsupervised Reenactment Algorithm to tackle these two challenges. Specifically, we propose to decompose the reenactment into three catenate processes: shape modeling, motion transfer and texture synthesis. With the decomposition, we introduce three crucial components, i.e., Parametric Shape Modeling, Expansionary Motion Transfer and Unsupervised Texture Synthesizer, to overcome the problems brought by the remarkably variances on pareidolia faces. Extensive experiments show the superior performance of our method both qualitatively and quantitatively. Code, model and data are available on our project page1.

Instance segmentation has witnessed a remarkable progress on class-balanced benchmarks. However, they fail to perform as accurately in real-world scenarios, where the category distribution of objects naturally comes with a long tail. Instances of head classes dominate a long-tailed dataset and they serve as negative samples of tail categories. The overwhelming gradients of negative samples on tail classes lead to a biased learning process for classifiers. Consequently, objects of tail categories are more likely to be misclassified as backgrounds or head categories. To tackle this problem, we propose Seesaw Loss to dynamically re-balance gradients of positive and negative samples for each category, with two complementary factors, i.e., mitigation factor and compensation factor. The mitigation factor reduces punishments to tail categories w.r.t. the ratio of cumulative training instances between different categories. Meanwhile, the compensation factor increases the penalty of misclassified instances to avoid false positives of tail categories. We conduct extensive experiments on Seesaw Loss with mainstream frameworks and different data sampling strategies. With a simple end-to-end training pipeline, Seesaw Loss obtains significant gains over Cross-Entropy Loss, and achieves state-of-the-art performance on LVIS dataset without bells and whistles. Code is available at https://github.com/open-mmlab/mmdetection.

SinGAN shows impressive capability in learning internal patch distribution despite its limited effective receptive field. We are interested in knowing how such a translationinvariant convolutional generator could capture the global structure with just a spatially i.i.d. input. In this work, taking SinGAN and StyleGAN2 as examples, we show that such capability, to a large extent, is brought by the implicit positional encoding when using zero padding in the generators. Such positional encoding is indispensable for generating images with high fidelity. The same phenomenon is observed in other generative architectures such as DCGAN and PGGAN. We further show that zero padding leads to an unbalanced spatial bias with a vague relation between locations. To offer a better spatial inductive bias, we investigate alternative positional encodings and analyze their effects. Based on a more flexible positional encoding explicitly, we propose a new multi-scale training strategy and demonstrate its effectiveness in the state-of-the-art unconditional generator StyleGAN2. Besides, the explicit spatial inductive bias substantially improves SinGAN for more versatile image manipulation.1

Most 3D shape completion approaches rely heavily on partial-complete shape pairs and learn in a fully super-vised manner. Despite their impressive performances on in-domain data, when generalizing to partial shapes in other forms or real-world partial scans, they often obtain unsatisfactory results due to domain gaps. In contrast to previous fully supervised approaches, in this paper we present ShapeInversion, which introduces Generative Adversarial Network (GAN) inversion to shape completion for the first time. ShapeInversion uses a GAN pre-trained on complete shapes by searching for a latent code that gives a complete shape that best reconstructs the given partial input. In this way, ShapeInversion no longer needs paired training data, and is capable of incorporating the rich prior captured in a well-trained generative model. On the ShapeNet bench-mark, the proposed ShapeInversion outperforms the SOTA unsupervised method, and is comparable with supervised methods that are learned using paired data. It also demonstrates remarkable generalization ability, giving robust results for real-world scans and partial inputs of various forms and incompleteness levels. Importantly, ShapeInversion naturally enables a series of additional abilities thanks to the involvement of a pre-trained GAN, such as producing multiple valid complete shapes for an ambiguous partial input, as well as shape manipulation and interpolation.

While accurate lip synchronization has been achieved for arbitrary-subject audio-driven talking face generation, the problem of how to efficiently drive the head pose remains. Previous methods rely on pre-estimated structural information such as landmarks and 3D parameters, aiming to generate personalized rhythmic movements. However, the inaccuracy of such estimated information under extreme conditions would lead to degradation problems. In this paper, we propose a clean yet effective framework to generate pose-controllable talking faces. We operate on non-aligned raw face images, using only a single photo as an identity reference. The key is to modularize audio-visual representations by devising an implicit low-dimension pose code. Substantially, both speech content and head pose information lie in a joint non-identity embedding space. While speech content information can be defined by learning the intrinsic synchronization between audio-visual modalities, we identify that a pose code will be complementarily learned in a modulated convolution-based reconstruction framework.Extensive experiments show that our method generates accurately lip-synced talking faces whose poses are controllable by other videos. Moreover, our model has multiple advanced capabilities including extreme view robustness and talking face frontalization.1

Facial editing is an important task in vision and graphics with numerous applications. However, existing works are incapable to deliver a continuous and fine-grained editing mode (e.g., editing a slightly smiling face to a big laughing one) with natural interactions with users. In this work, we propose Talk-to-Edit, an interactive facial editing framework that performs fine-grained attribute manipulation through dialog between the user and the system. Our key insight is to model a continual "semantic field" in the GAN latent space. 1) Unlike previous works that regard the editing as traversing straight lines in the latent space, here the fine-grained editing is formulated as finding a curving trajectory that respects fine-grained attribute landscape on the semantic field. 2) The curvature at each step is location-specific and determined by the input image as well as the users’ language requests. 3) To engage the users in a meaningful dialog, our system generates language feedback by considering both the user request and the current state of the semantic field.We also contribute CelebA-Dialog, a visual-language facial editing dataset to facilitate large-scale study. Specifically, each image has manually annotated fine-grained attribute annotations as well as template-based textual descriptions in natural language. Extensive quantitative and qualitative experiments demonstrate the superiority of our framework in terms of 1) the smoothness of fine-grained editing, 2) the identity/attribute preservation, and 3) the visual photorealism and dialog fluency. Notably, user study validates that our overall system is consistently favored by around 80% of the participants. Our project page is https://www.mmlab-ntu.com/project/talkedit/.

Image reconstruction and synthesis have witnessed remarkable progress thanks to the development of generative models. Nonetheless, gaps could still exist between the real and generated images, especially in the frequency domain. In this study, we show that narrowing gaps in the frequency domain can ameliorate image reconstruction and synthesis quality further. We propose a novel focal frequency loss, which allows a model to adaptively focus on frequency components that are hard to synthesize by down-weighting the easy ones. This objective function is complementary to existing spatial losses, offering great impedance against the loss of important frequency information due to the inherent bias of neural networks. We demonstrate the versatility and effectiveness of focal frequency loss to improve popular models, such as VAE, pix2pix, and SPADE, in both perceptual quality and quantitative performance. We further show its potential on StyleGAN2.1,2

We propose a Transformer-based framework for 3D human texture estimation from a single image. The proposed Transformer is able to effectively exploit the global information of the input image, overcoming the limitations of existing methods that are solely based on convolutional neural networks. In addition, we also propose a mask-fusion strategy to combine the advantages of the RGB-based and texture-flow-based models. We further introduce a part-style loss to help reconstruct high-fidelity colors without introducing unpleasant artifacts. Extensive experiments demonstrate the effectiveness of the proposed method against state-of-the-art 3D human texture estimation approaches both quantitatively and qualitatively. The project page is at https://www.mmlab-ntu.com/project/texformer.

Image Signal Processor (ISP) is a crucial component in digital cameras that transforms sensor signals into images for us to perceive and understand. Existing ISP designs always adopt a fixed architecture, e.g., several sequential modules connected in a rigid order. Such a fixed ISP architecture may be suboptimal for real-world applications, where camera sensors, scenes and tasks are diverse. In this study, we propose a novel Reconfigurable ISP (ReconfigISP) whose architecture and parameters can be automatically tailored to specific data and tasks. In particular, we implement several ISP modules, and enable back-propagation for each module by training a differentiable proxy, hence allowing us to leverage the popular differentiable neural architecture search and effectively search for the optimal ISP architecture. A proxy tuning mechanism is adopted to maintain the accuracy of proxy networks in all cases. Extensive experiments conducted on image restoration and object detection, with different sensors, light conditions and efficiency constraints, validate the effectiveness of ReconfigISP. Only hundreds of parameters need tuning for every task1.

Recent methods for long-tailed instance segmentation still struggle on rare object classes with few training data. We propose a simple yet effective method, Feature Augmentation and Sampling Adaptation (FASA), that addresses the data scarcity issue by augmenting the feature space especially for rare classes. Both the Feature Augmentation (FA) and feature sampling components are adaptive to the actual training status — FA is informed by the feature mean and variance of observed real samples from past iterations, and we sample the generated virtual features in a loss-adapted manner to avoid over-fitting. FASA does not require any elaborate loss design, and removes the need for inter-class transfer learning that often involves large cost and manually-defined head/tail class groups. We show FASA is a fast, generic method that can be easily plugged into standard or long-tailed segmentation frameworks, with consistent performance gains and little added cost. FASA is also applicable to other tasks like long-tailed classification with state-of-the-art performance.12

Existing works study the Class Incremental learning (CIL) problem with the assumption that the data for previous classes are absent, or only a small subset of samples (known as exemplars) are accessible. Differently, we propose a new and practical setting called retrospective CIL, where all the previous data are accessible, but with bounded training budgets for old data replay. Since only a small subset of old samples can be replayed, it brings a new research problem, i.e., dynamically sampling old data along the incremental training process. As incremental learning particularly suffers from catastrophic forgetting, we propose to use the forgettability of the old samples as the sampling priorities to favour the forgotten samples during the dynamic sampling process. To achieve this, we introduce a forgetting rate metric with graph- based propagation to estimate the sample forgettability. The proposed method brings improvements on two benchmark datasets.

Generative adversarial networks (GANs) typically require ample data for training in order to synthesize high-fidelity images. Recent studies have shown that training GANs with limited data remains formidable due to discriminator overfitting, the underlying cause that impedes the generator's convergence. This paper introduces a novel strategy called Adaptive Pseudo Augmentation (APA) to encourage healthy competition between the generator and the discriminator. As an alternative method to existing approaches that rely on standard data augmentations or model regularization, APA alleviates overfitting by employing the generator itself to augment the real data distribution with generated images, which deceives the discriminator adaptively. Extensive experiments demonstrate the effectiveness of APA in improving synthesis quality in the low-data regime. We provide a theoretical analysis to examine the convergence and rationality of our new training strategy. APA is simple and effective. It can be added seamlessly to powerful contemporary GANs, such as StyleGAN2, with negligible computational cost. Code: https://github.com/EndlessSora/DeceiveD.

The advancement of generative radiance fields has pushed the boundary of 3D-aware image synthesis. Motivated by the observation that a 3D object should look realistic from multiple viewpoints, these methods introduce a multi-view constraint as regularization to learn valid 3D radiance fields from 2D images. Despite the progress, they often fall short of capturing accurate 3D shapes due to the shape-color ambiguity, limiting their applicability in downstream tasks. In this work, we address this ambiguity by proposing a novel shading-guided generative implicit model that is able to learn a starkly improved shape representation. Our key insight is that an accurate 3D shape should also yield a realistic rendering under different lighting conditions. This multi-lighting constraint is realized by modeling illumination explicitly and performing shading with various lighting conditions. Gradients are derived by feeding the synthesized images to a discriminator. To compensate for the additional computational burden of calculating surface normals, we further devise an efficient volume rendering strategy via surface tracking, reducing the training and inference time by 24% and 48%, respectively. Our experiments on multiple datasets show that the proposed approach achieves photorealistic 3D-aware image synthesis while capturing accurate underlying 3D shapes. We demonstrate improved performance of our approach on 3D shape reconstruction against existing methods, and show its applicability on image relighting. Our code is available at https://github.com/XingangPan/ShadeGAN.

Contrastive self-supervised learning has largely narrowed the gap to supervised pre-training on ImageNet. However, its success highly relies on the object-centric priors of ImageNet, i.e., different augmented views of the same image correspond to the same object. Such a heavily curated constraint becomes immediately infeasible when pre-trained on more complex scene images with many objects. To overcome this limitation, we introduce Object-level Representation Learning (ORL), a new self-supervised learning framework towards scene images. Our key insight is to leverage image-level self-supervised pre-training as the prior to discover object-level semantic correspondence, thus realizing object-level representation learning from scene images. Extensive experiments on COCO show that ORL significantly improves the performance of self-supervised learning on scene images, even surpassing supervised ImageNet pre-training on several downstream tasks. Furthermore, ORL improves the downstream performance when more unlabeled scene images are available, demonstrating its great potential of harnessing unlabeled data in the wild. We hope our approach can motivate future research on more general-purpose unsupervised representation learning from scene data.

Semantic, instance, and panoptic segmentations have been addressed using different and specialized frameworks despite their underlying connections. This paper presents a unified, simple, and effective framework for these essentially similar tasks. The framework, named K-Net, segments both instances and semantic categories consistently by a group of learnable kernels, where each kernel is responsible for generating a mask for either a potential instance or a stuff class. To remedy the difficulties of distinguishing various instances, we propose a kernel update strategy that enables each kernel dynamic and conditional on its meaningful group in the input image. K-Net can be trained in an end-to-end manner with bipartite matching, and its training and inference are naturally NMS-free and box-free. Without bells and whistles, K-Net surpasses all previous published state-of-the-art single-model results of panoptic segmentation on MS COCO test-dev split and semantic segmentation on ADE20K val split with 55.2% PQ and 54.3% mIoU, respectively. Its instance segmentation performance is also on par with Cascade Mask R-CNN on MS COCO with 60%-90% faster inference speeds. Code and models will be released at https://github.com/ZwwWayne/K-Net/.

Over four decades, the majority addresses the problem of optical flow estimation using variational methods. With the advance of machine learning, some recent works have attempted to address the problem using convolutional neural network (CNN) and have showed promising results. FlowNet2 [1] , the state-of-the-art CNN, requires over 160M parameters to achieve accurate flow estimation. Our LiteFlowNet2 outperforms FlowNet2 on Sintel and KITTI benchmarks, while being 25.3 times smaller in the model size and 3.1 times faster in the running speed. LiteFlowNet2 is built on the foundation laid by conventional methods and resembles the corresponding roles as data fidelity and regularization in variational methods. We compute optical flow in a spatial-pyramid formulation as SPyNet [2] but through a novel lightweight cascaded flow inference. It provides high flow estimation accuracy through early correction with seamless incorporation of descriptor matching. Flow regularization is used to ameliorate the issue of outliers and vague flow boundaries through feature-driven local convolutions. Our network also owns an effective structure for pyramidal feature extraction and embraces feature warping rather than image warping as practiced in FlowNet2 and SPyNet. Comparing to LiteFlowNet [3] , LiteFlowNet2 improves the optical flow accuracy on Sintel Clean by 23.3 percent, Sintel Final by 12.8 percent, KITTI 2012 by 19.6 percent, and KITTI 2015 by 18.8 percent, while being 2.2 times faster. Our network protocol and trained models are made publicly available on https://github.com/twhui/LiteFlowNet2.

Patch-based methods and deep networks have been employed to tackle image inpainting problem, with their own strengths and weaknesses. Patch-based methods are capable of restoring a missing region with high-quality texture through searching nearest neighbor patches from the unmasked regions. However, these methods bring problematic contents when recovering large missing regions. Deep networks, on the other hand, show promising results in completing large regions. Nonetheless, the results often lack faithful and sharp details that resemble the surrounding area. By bringing together the best of both paradigms, we propose a new deep inpainting framework where texture generation is guided by a texture memory of patch samples extracted from unmasked regions. The framework has a novel design that allows texture memory retrieval to be trained end-to-end with the deep inpainting network. In addition, we introduce a patch distribution loss to encourage high-quality patch synthesis. The proposed method shows superior performance both qualitatively and quantitatively on three challenging image benchmarks, i.e., Places, CelebA-HQ, and Paris Street-View datasets (Code will be made publicly available in https://github.com/open-mmlab/mmediting).

It is a common paradigm in object detection frameworks to treat all samples equally and target at maximizing the performance on average. In this work, we revisit this paradigm through a careful study on how different samples contribute to the overall performance measured in terms of mAP. Our study suggests that the samples in each mini-batch are neither independent nor equally important, and therefore a better classifier on average does not necessarily result in higher mAP. Motivated by this study, we propose the notion of Prime Samples, those that play a key role in driving the detection performance. We further develop a simple yet effective sampling and learning strategy called PrIme Sample Attention (PISA) that directs the focus of the training process towards such samples. Our experiments demonstrate that it is often more effective to focus on prime samples than hard samples when training a detector. Particularly, on the MSCOCO dataset, PISA outperforms the random sampling baseline and hard mining schemes, \eg~OHEM and Focal Loss, consistently by around 2\% on both single-stage and two-stage detectors, even with a strong backbone ResNeXt-101. Code is available at: \url{https://github.com/open-mmlab/mmdetection}.

The paper presents a novel method, Zero-Reference Deep Curve Estimation (Zero-DCE), which formulates light enhancement as a task of image-specific curve estimation with a deep network. Our method trains a lightweight deep network, DCE-Net, to estimate pixel-wise and high-order curves for dynamic range adjustment of a given image. The curve estimation is specially designed, considering pixel value range, monotonicity, and differentiability. Zero-DCE is appealing in its relaxed assumption on reference images, i.e., it does not require any paired or unpaired data during training. This is achieved through a set of carefully formulated non-reference loss functions, which implicitly measure the enhancement quality and drive the learning of the network. Our method is efficient as image enhancement can be achieved by an intuitive and simple nonlinear curve mapping. Despite its simplicity, we show that it generalizes well to diverse lighting conditions. Extensive experiments on various benchmarks demonstrate the advantages of our method over state-of-the-art methods qualitatively and quantitatively. Furthermore, the potential benefits of our Zero-DCE to face detection in the dark are discussed.

We study the problem of distilling knowledge from a large deep teacher network to a much smaller student network for the task of road marking segmentation. In this work, we explore a novel knowledge distillation (KD) approach that can transfer `knowledge' on scene structure more effectively from a teacher to a student model. Our method is known as Inter-Region Affinity KD (IntRA-KD). It decomposes a given road scene image into different regions and represents each region as a node in a graph. An inter-region affinity graph is then formed by establishing pairwise relationships between nodes based on their similarity in feature distribution. To learn structural knowledge from the teacher network, the student is required to match the graph generated by the teacher. The proposed method shows promising results on three large-scale road marking segmentation benchmarks, i.e., ApolloScape, CULane and LLAMAS, by taking various lightweight models as students and ResNet-101 as the teacher. IntRA-KD consistently brings higher performance gains on all lightweight models, compared to previous distillation methods. Our code is available at https://github.com/ cardwing/Codes-for-IntRA-KD.

We present our on-going effort of constructing a large- scale benchmark for face forgery detection. The first version of this benchmark, DeeperForensics-1.0, represents the largest face forgery detection dataset by far, with 60, 000 videos constituted by a total of 17.6 million frames, 10 times larger than existing datasets of the same kind. Extensive real-world perturbations are applied to obtain a more challenging benchmark of larger scale and higher diversity. All source videos in DeeperForensics-1.0 are carefully collected, and fake videos are generated by a newly proposed end-to-end face swapping framework. The quality of generated videos outperforms those in existing datasets, validated by user studies. The benchmark features a hidden test set, which contains manipulated videos achieving high deceptive scores in human evaluations. We further contribute a comprehensive study that evaluates five representative detection baselines and make a thorough analysis of different settings.

Face clustering is an essential tool for exploiting the unlabeled face data, and has a wide range of applications including face annotation and retrieval. Recent works show that supervised clustering can result in noticeable performance gain. However, they usually involve heuristic steps and require numerous overlapped subgraphs, severely restricting their accuracy and efficiency. In this paper, we propose a fully learnable clustering framework without requiring a large number of overlapped subgraphs. Instead, we transform the clustering problem into two sub-problems. Specifically, two graph convolutional networks, named GCN-V and GCN-E, are designed to estimate the confidence of vertices and the connectivity of edges, respectively. With the vertex confidence and edge connectivity, we can naturally organize more relevant vertices on the affinity graph and group them into clusters. Experiments on two large-scale benchmarks show that our method significantly improves clustering accuracy and thus performance of the recognition models trained on top, yet it is an order of magnitude more efficient than existing supervised methods.

We present a lightweight video motion retargeting approach TransMoMo that is capable of transferring motion of a person in a source video realistically to another video of a target person. Without using any paired data for supervision, the proposed method can be trained in an unsupervised manner by exploiting invariance properties of three orthogonal factors of variation including motion, structure, and view-angle. Specifically, with loss functions carefully derived based on invariance, we train an auto-encoder to disentangle the latent representations of such factors given the source and target video clips. This allows us to selectively transfer motion extracted from the source video seamlessly to the target video in spite of structural and view-angle disparities between the source and the target. The relaxed assumption of paired data allows our method to be trained on a vast amount of videos needless of manual annotation of source-target pairing, leading to improved robustness against large structural variations and extreme motion in videos. We demonstrate the effectiveness of our method over the state-of-the-art methods. Code, model and data are publicly available on our project page (https://yzhq97.github.io/transmomo).

Natural scene understanding is a challenging task, particularly when encountering images of multiple objects that are partially occluded. This obstacle is given rise by varying object ordering and positioning. Existing scene understanding paradigms are able to parse only the visible parts, resulting in incomplete and unstructured scene interpretation. In this paper, we investigate the problem of scene de-occlusion, which aims to recover the underlying occlusion ordering and complete the invisible parts of occluded objects. We make the first attempt to address the problem through a novel and unified framework that recovers hidden scene structures without ordering and amodal annotations as supervisions. This is achieved via Partial Completion Network (PCNet)-mask (M) and -content (C), that learn to recover fractions of object masks and contents, respectively, in a self-supervised manner. Based on PCNet-M and PCNet-C, we devise a novel inference scheme to accomplish scene de-occlusion, via progressive ordering recovery, amodal completion and content completion. Extensive experiments on real-world scenes demonstrate the superior performance of our approach to other alternatives. Remarkably, our approach that is trained in a self-supervised manner achieves comparable results to fully-supervised methods. The proposed scene de-occlusion framework benefits many applications, including high-quality and controllable image manipulation and scene recomposition (see Fig. 1), as well as the conversion of existing modal mask annotations to amodal mask annotations.

Joint clustering and feature learning methods have shown remarkable performance in unsupervised representation learning. However, the training schedule alternating between feature clustering and network parameters update leads to unstable learning of visual representations. To overcome this challenge, we propose Online Deep Clustering (ODC) that performs clustering and network update simultaneously rather than alternatingly. Our key insight is that the cluster centroids should evolve steadily in keeping the classifier stably updated. Specifically, we design and maintain two dynamic memory modules, i.e., samples memory to store samples' labels and features, and centroids memory for centroids evolution. We break down the abrupt global clustering into steady memory update and batch-wise label re-assignment. The process is integrated into network update iterations. In this way, labels and the network evolve shoulder-to-shoulder rather than alternatingly. Extensive experiments demonstrate that ODC stabilizes the training process and boosts the performance effectively.

Neural Architecture Search (NAS) achieves significant progress in many computer vision tasks. While many methods are proposed to improve the efficiency of NAS, the search progress is still laborious because training and evaluating plausible architectures over large search space is time-consuming. Assessing network candidates under a proxy (i.e., computationally reduced setting) thus becomes inevitable. In this paper, we observe that most existing proxies exhibit different behaviors in maintaining the rank consistency among network candidates. In particular, some proxies can be more reliable - the rank of candidates does not differ much comparing their reduced setting performance and final performance. In this paper, we systematically investigate some widely adopted reduction factors and report our observations. Inspired by these observations, we present a reliable proxy and further formulate a hierarchical proxy strategy that spends more computations on candidate networks that are potentially more accurate, while discards unpromising ones in early stage with a fast proxy. This leads to an economical evolutionary-based NAS (EcoNAS), which achieves an impressive 400×search time reduction in comparison to the evolutionary-based state of the art [19] (8 v.s. 3150 GPU days). Some new proxies led by our observations can also be applied to accelerate other NAS methods while still able to discover good candidate networks with performance matching those found by previous proxy strategies. Codes and models will be released to facilitate future research.

Non-local self-similarity in natural images has been well studied as an effective prior in image restoration. However, for single image super-resolution (SISR), most existing deep non-local methods (e.g., non-local neural networks) only exploit similar patches within the same scale of the low-resolution (LR) input image. Consequently, the restoration is limited to using the same-scale information while neglecting potential high-resolution (HR) cues from other scales. In this paper, we explore the cross-scale patch recurrence property of a natural image, i.e., similar patches tend to recur many times across different scales. This is achieved using a novel cross-scale internal graph neural network (IGNN). Specifically, we dynamically construct a cross-scale graph by searching k-nearest neighboring patches in the downsampled LR image for each query patch in the LR image. We then obtain the corresponding k HR neighboring patches in the LR image and aggregate them adaptively in accordance to the edge label of the constructed graph. In this way, the HR information can be passed from k HR neighboring patches to the LR query patch to help it recover more detailed textures. Besides, these internal image-specific LR/HR exemplars are also significant complements to the external information learned from the training dataset. Extensive experiments demonstrate the effectiveness of IGNN against the state-of-the-art SISR methods including existing non-local networks on standard benchmarks.

Data for face analysis often exhibit highly-skewed class distribution, i.e., most data belong to a few majority classes, while the minority classes only contain a scarce amount of instances. To mitigate this issue, contemporary deep learning methods typically follow classic strategies such as class re-sampling or cost-sensitive training. In this paper, we conduct extensive and systematic experiments to validate the effectiveness of these classic schemes for representation learning on class-imbalanced data. We further demonstrate that more discriminative deep representation can be learned by enforcing a deep network to maintain inter-cluster margins both within and between classes. This tight constraint effectively reduces the class imbalance inherent in the local data neighborhood, thus carving much more balanced class boundaries locally. We show that it is easy to deploy angular margins between the cluster distributions on a hypersphere manifold. Such learned Cluster-based Large Margin Local Embedding (CLMLE), when combined with a simple k-nearest cluster algorithm, shows significant improvements in accuracy over existing methods on both face recognition and face attribute prediction tasks that exhibit imbalanced class distribution.

The training of many existing end-to-end steering angle prediction models heavily relies on steering angles as the supervisory signal. Without learning from much richer contexts, these methods are susceptible to the presence of sharp road curves, challenging traffic conditions, strong shadows, and severe lighting changes. In this paper, we considerably improve the accuracy and robustness of predictions through heterogeneous auxiliary networks feature mimicking, a new and effective training method that provides us with much richer contextual signals apart from steering direction. Specifically, we train our steering angle predictive model by distilling multi-layer knowledge from multiple heterogeneous auxiliary networks that perform related but different tasks, e.g., image segmentation or optical flow estimation. As opposed to multi-task learning, our method does not require expensive annotations of related tasks on the target set. This is made possible by applying contemporary off-the-shelf networks on the target set and mimicking their features in different layers after transformation. The auxiliary networks are discarded after training without affecting the runtime efficiency of our model. Our approach achieves a new state-of-the-art on Udacity and Comma.ai, outperforming the previous best by a large margin of 12.8% and 52.1%1, respectively. Encouraging results are also shown on Berkeley Deep Drive (BDD) dataset.

We address the problem of instance-level facial attribute transfer without paired training data, e.g., faithfully transferring the exact mustache from a source face to a target face. This is a more challenging task than the conventional semantic-level attribute transfer, which only preserves the generic attribute style instead of instance-level traits. We propose the use of geometry-aware flow, which serves as a wellsuited representation for modeling the transformation between instance-level facial attributes. Specifically, we leverage the facial landmarks as the geometric guidance to learn the differentiable flows automatically, despite of the large pose gap existed. Geometry-aware flow is able to warp the source face attribute into the target face context and generate a warp-and-blend result. To compensate for the potential appearance gap between source and target faces, we propose a hallucination sub-network that produces an appearance residual to further refine the warp-and-blend result. Finally, a cycle-consistency framework consisting of both attribute transfer module and attribute removal module is designed, so that abundant unpaired face images can be used as training data. Extensive evaluations validate the capability of our approach in transferring instance-level facial attributes faithfully across large pose and appearance gaps. Thanks to the flow representation, our approach can readily be applied to generate realistic details on high-resolution images1.

Cascade is a classic yet powerful architecture that has boosted performance on various tasks. However, how to introduce cascade to instance segmentation remains an open question. A simple combination of Cascade R-CNN and Mask R-CNN only brings limited gain. In exploring a more effective approach, we find that the key to a successful instance segmentation cascade is to fully leverage the reciprocal relationship between detection and segmentation. In this work, we propose a new framework, Hybrid Task Cascade (HTC), which differs in two important aspects: (1) instead of performing cascaded refinement on these two tasks separately, it interweaves them for a joint multi-stage processing; (2) it adopts a fully convolutional branch to provide spatial context, which can help distinguishing hard foreground from cluttered background. Overall, this framework can learn more discriminative features progressively while integrating complementary features together in each stage. Without bells and whistles, a single HTC obtains 38.4% and 1.5% improvement over a strong Cascade Mask R-CNN baseline on MSCOCO dataset. Moreover, our overall system achieves 48.6 mask AP on the test-challenge split, ranking 1st in the COCO 2018 Challenge Object Detection Task. Code is available at https://github.com/open-mmlab/mmdetection.

Conventionally, deep neural networks are trained offline, relying on a large dataset prepared in advance. This paradigm is often challenged in real-world applications, e.g. online services that involve continuous streams of incoming data. Recently, incremental learning receives increasing attention, and is considered as a promising solution to the practical challenges mentioned above. However, it has been observed that incremental learning is subject to a fundamental difficulty -- catastrophic forgetting, namely adapting a model to new data often results in severe performance degradation on previous tasks or classes. Our study reveals that the imbalance between previous and new data is a crucial cause to this problem. In this work, we develop a new framework for incrementally learning a unified classifier, e.g. a classifier that treats both old and new classes uniformly. Specifically, we incorporate three components, cosine normalization, less-forget constraint, and inter-class separation, to mitigate the adverse effects of the imbalance. Experiments show that the proposed method can effectively rebalance the training process, thus obtaining superior performance compared to the existing methods. On CIFAR-100 and ImageNet, our method can reduce the classification errors by more than 6% and 13% respectively, under the incremental setting of 10 phases.

We present a novel approach for the task of human pose transfer, which aims at synthesizing a new image of a person from an input image of that person and a target pose. We address the issues of limited correspondences identified between keypoints only and invisible pixels due to self-occlusion. Unlike existing methods, we propose to estimate dense and intrinsic 3D appearance flow to better guide the transfer of pixels between poses. In particular, we wish to generate the 3D flow from just the reference and target poses. Training a network for this purpose is non-trivial, especially when the annotations for 3D appearance flow are scarce by nature. We address this problem through a flow synthesis stage. This is achieved by fitting a 3D model to the given pose pair and project them back to the 2D plane to compute the dense appearance flow for training. The synthesized ground-truths are then used to train a feedforward network for efficient mapping from the input and target skeleton poses to the 3D appearance flow. With the appearance flow, we perform feature warping on the input image and generate a photorealistic image of the target pose. Extensive results on DeepFashion and Market-1501 datasets demonstrate the effectiveness of our approach over existing methods. Our code is available at http://mmlab.ie.cuhk.edu.hk/projects/pose-transfer.

Region anchors are the cornerstone of modern object detection techniques. State-of-the-art detectors mostly rely on a dense anchoring scheme, where anchors are sampled uniformly over the spatial domain with a predefined set of scales and aspect ratios. In this paper, we revisit this foundational stage. Our study shows that it can be done much more effectively and efficiently. Specifically, we present an alternative scheme, named Guided Anchoring, which leverages semantic features to guide the anchoring. The proposed method jointly predicts the locations where the center of objects of interest are likely to exist as well as the scales and aspect ratios at different locations. On top of predicted anchor shapes, we mitigate the feature inconsistency with a feature adaption module. We also study the use of high-quality proposals to improve detection performance. The anchoring scheme can be seamlessly integrated into proposal methods and detectors. With Guided Anchoring, we achieve 9.1% higher recall on MS COCO with 90% fewer anchors than the RPN baseline. We also adopt Guided Anchoring in Fast R-CNN, Faster R-CNN and RetinaNet, respectively improving the detection mAP by 2.2%, 2.7% and 1.2%. Code is available at https://github.com/open-mmlab/mmdetection.

Deep convolutional neural network has demonstrated its capability of learning a deterministic mapping for the desired imagery effect. However, the large variety of user flavors motivates the possibility of continuous transition among different output effects. Unlike existing methods that require a specific design to achieve one particular transition (e.g., style transfer), we propose a simple yet universal approach to attain a smooth control of diverse imagery effects in many low-level vision tasks, including image restoration, image-to-image translation, and style transfer. Specifically, our method, namely Deep Network Interpolation (DNI), applies linear interpolation in the parameter space of two or more correlated networks. A smooth control of imagery effects can be achieved by tweaking the interpolation coefficients. In addition to DNI and its broad applications, we also investigate the mechanism of network interpolation from the perspective of learned filters.

Unsupervised image-to-image translation aims at learning a mapping between two visual domains. However, learning a translation across large geometry variations al- ways ends up with failure. In this work, we present a novel disentangle-and-translate framework to tackle the complex objects image-to-image translation task. Instead of learning the mapping on the image space directly, we disentangle image space into a Cartesian product of the appearance and the geometry latent spaces. Specifically, we first in- troduce a geometry prior loss and a conditional VAE loss to encourage the network to learn independent but com- plementary representations. The translation is then built on appearance and geometry space separately. Extensive experiments demonstrate the superior performance of our method to other state-of-the-art approaches, especially in the challenging near-rigid and non-rigid objects translation tasks. In addition, by taking different exemplars as the ap- pearance references, our method also supports multimodal translation. Project page: https://wywu.github. io/projects/TGaGa/TGaGa.html

Video inpainting, which aims at filling in missing regions in a video, remains challenging due to the difficulty of preserving the precise spatial and temporal coherence of video contents. In this work we propose a novel flow-guided video inpainting approach. Rather than filling in the RGB pixels of each frame directly, we consider the video inpainting as a pixel propagation problem. We first synthesize a spatially and temporally coherent optical flow field across video frames using a newly designed Deep Flow Completion network, then use the synthesized flow fields to guide the propagation of pixels to fill up the missing regions in the video. Specifically, the Deep Flow Competion network follows a coarse-to-fine refinement strategy to complete the flow fields, while their quality is further improved by hard flow example mining. Following the guide of the completed flow fields, the missing video regions can be filled up precisely. Our method is evaluated on DAVIS and YouTubeVOS datasets qualitatively and quantitatively, achieving the state-of-the-art performance in terms of inpainting quality and speed.

Face recognition sees remarkable progress in recent years, and its performance has reached a very high level. Taking it to a next level requires substantially larger data, which would involve prohibitive annotation cost. Hence, exploiting unlabeled data becomes an appealing alternative. Recent works have shown that clustering unlabeled faces is a promising approach, often leading to notable performance gains. Yet, how to effectively cluster, especially on a large-scale (i.e. million-level or above) dataset, remains an open question. A key challenge lies in the complex variations of cluster patterns, which make it difficult for conventional clustering methods to meet the needed accuracy. This work explores a novel approach, namely, learning to cluster instead of relying on hand-crafted criteria. Specifically, we propose a framework based on graph convolutional network, which combines a detection and a segmentation module to pinpoint face clusters. Experiments show that our method yields significantly more accurate face clusters, which, as a result, also lead to further performance gain in face recognition.

Intelligent agent naturally learns from motion. Various self-supervised algorithms have leveraged the motion cues to learn effective visual representations. The hurdle here is that motion is both ambiguous and complex, rendering previous works either suffer from degraded learning efficacy, or resort to strong assumptions on object motions. In this work, we design a new learning-from-motion paradigm to bridge these gaps. Instead of explicitly modeling the motion probabilities, we design the pretext task as a conditional motion propagation problem. Given an input image and several sparse flow guidance on it, our framework seeks to recover the full-image motion. Compared to other alternatives, our framework has several appealing properties: (1) Using sparse flow guidance during training resolves the inherent motion ambiguity, and thus easing feature learning. (2) Solving the pretext task of conditional motion propagation encourages the emergence of kinematically-sound representations that poss greater expressive power. Extensive experiments demonstrate that our framework learns structural and coherent features; and achieves state-of-the-art self-supervision performance on several downstream tasks including semantic segmentation, instance segmentation and human parsing. Furthermore, our framework is successfully extended to several useful applications such as semi-automatic pixel-level annotation.

Training deep models for lane detection is challenging due to the very subtle and sparse supervisory signals inherent in lane annotations. Without learning from much richer context, these models often fail in challenging scenarios, e.g., severe occlusion, ambiguous lanes, and poor lighting conditions. In this paper, we present a novel knowledge distillation approach, i.e., Self Attention Distillation (SAD), which allows a model to learn from itself and gains substantial improvement without any additional supervision or labels. Specifically, we observe that attention maps extracted from a model trained to a reasonable level would encode rich contextual information. The valuable contextual information can be used as a form of `free' supervision for further representation learning through performing top- down and layer-wise attention distillation within the net- work itself. SAD can be easily incorporated in any feed- forward convolutional neural networks (CNN) and does not increase the inference time. We validate SAD on three popular lane detection benchmarks (TuSimple, CULane and BDD100K) using lightweight models such as ENet, ResNet- 18 and ResNet-34. The lightest model, ENet-SAD, performs comparatively or even surpasses existing algorithms. Notably, ENet-SAD has 20 × fewer parameters and runs 10 × faster compared to the state-of-the-art SCNN, while still achieving compelling performance in all benchmarks.

Though much progress has been achieved in single-image 3D human recovery, estimating 3D model for in-the-wild images remains a formidable challenge. The reason lies in the fact that obtaining high-quality 3D annotations for in-the-wild images is an extremely hard task that consumes enormous amount of resources and manpower. To tackle this problem, previous methods adopt a hybrid training strategy that exploits multiple heterogeneous types of annotations including 3D and 2D while leaving the efficacy of each annotation not thoroughly investigated. In this work, we aim to perform a comprehensive study on cost and effectiveness trade-off between different annotations. Specifically, we focus on the challenging task of in-the-wild 3D human recovery from single images when paired 3D annotations are not fully available. Through extensive experiments, we obtain several observations: 1) 3D annotations are efficient, whereas traditional 2D annotations such as 2D keypoints and body part segmentation are less competent in guiding 3D human recovery. 2) Dense Correspondence such as DensePose is effective. When there are no paired in-the-wild 3D annotations available, the model exploiting dense correspondence can achieve 92% of the performance compared to a model trained with paired 3D data. We show that incorporating dense correspondence into in-the-wild 3D human recovery is promising and competitive due to its high efficiency and relatively low annotating cost. Our model trained with dense correspondence can serve as a strong reference for future research.

Feature upsampling is a key operation in a number of modern convolutional network architectures, e.g. feature pyramids. Its design is critical for dense prediction tasks such as object detection and semantic/instance segmentation. In this work, we propose Content-Aware ReAssembly of FEatures (CARAFE), a universal, lightweight and highly effective operator to fulfill this goal. CARAFE has several appealing properties: (1) Large field of view. Unlike previous works (e.g. bilinear interpolation) that only exploit subpixel neighborhood, CARAFE can aggregate contextual information within a large receptive field. (2) Content-aware handling. Instead of using a fixed kernel for all samples (e.g. deconvolution), CARAFE enables instance-specific content-aware handling, which generates adaptive kernels on-the-fly. (3) Lightweight and fast to compute. CARAFE introduces little computational overhead and can be readily integrated into modern network architectures. We conduct comprehensive evaluations on standard benchmarks in object detection, instance/semantic segmentation and inpainting. CARAFE shows consistent and substantial gains across all the tasks (1.2% AP, 1.3% AP, 1.8% mIoU, 1.1dB respectively) with negligible computational overhead. It has great potential to serve as a strong building block for future research. Code and models are available at https://github.com/open-mmlab/mmdetection.

Multi-sensor perception is crucial to ensure the reliability and accuracy in autonomous driving system, while multi-object tracking (MOT) improves that by tracing sequential movement of dynamic objects. Most current approaches for multi-sensor multi-object tracking are either lack of reliability by tightly relying on a single input source (e.g., center camera), or not accurate enough by fusing the results from multiple sensors in post processing without fully exploiting the inherent information. In this study, we design a generic sensor-agnostic multi-modality MOT framework (mmMOT), where each modality (i.e., sensors) is capable of performing its role independently to preserve reliability, and could further improving its accuracy through a novel multi-modality fusion module. Our mmMOT can be trained in an end-to-end manner, enables joint optimization for the base feature extractor of each modality and an adjacency estimator for cross modality. Our mmMOT also makes the first attempt to encode deep representation of point cloud in data association process in MOT. We conduct extensive experiments to evaluate the effectiveness of the proposed framework on the challenging KITTI benchmark and report state-of-the-art performance. Code and models are available at https://github.com/ZwwWayne/mmMOT.

Face grouping remains a challenging problem despite the remarkable capability of deep learning approaches in learning face representation. In particular, grouping results can still be egregious given profile faces and a large number of uninteresting faces and noisy detections. Often, a user needs to correct the erroneous grouping manually. In this study, we formulate a novel face grouping framework that learns clustering strategy from ground-truth simulated behavior. This is achieved through imitation learning (a.k.a apprenticeship learning or learning by watching) via inverse reinforcement learning (IRL). In contrast to existing clustering approaches that group instances by similarity, our framework makes sequential decision to dynamically decide when to merge two face instances/groups driven by short- and long-term rewards. Extensive experiments on three benchmark datasets show that our framework outperforms unsupervised and supervised baselines.

Deep convolutional networks for semantic image segmentation typically require large-scale labeled data, e.g., ImageNet and MS COCO, for network pre-training. To reduce annotation efforts, self-supervised semantic segmentation is recently proposed to pre-train a network without any human-provided labels. The key of this new form of learning is to design a proxy task (e.g., image colorization), from which a discriminative loss can be formulated on unlabeled data. Many proxy tasks, however, lack the critical supervision signals that could induce discriminative representation for the target image segmentation task. Thus self-supervision’s performance is still far from that of supervised pre-training. In this study, we overcome this limitation by incorporating a "mix-and-match" (M&M) tuning stage in the self-supervision pipeline. The proposed approach is readily pluggable to many self-supervision methods and does not use more annotated samples than the original process. Yet, it is capable of boosting the performance of target image segmentation task to surpass fully-supervised pre-trained counterpart. The improvement is made possible by better harnessing the limited pixel-wise annotations in the target dataset. Specifically, we first introduce the "mix" stage, which sparsely samples and mixes patches from the target set to reflect rich and diverse local patch statistics of target images. A ‘match’ stage then forms a class-wise connected graph, which can be used to derive a strong triplet-based discriminative loss for finetuning the network. Our paradigm follows the standard practice in existing self-supervised studies and no extra data or label is required. With the proposed M&M approach, for the first time, a self-supervision method can achieve comparable or even better performance compared to its ImageNet pretrained counterpart on both PASCAL VOC2012 dataset and CityScapes dataset.

Face recognition achieves exceptional success thanks to the emergence of deep learning. However, many contemporary face recognition models still perform relatively poor in processing profile faces compared to frontal faces. A key reason is that the number of frontal and profile training faces are highly imbalanced - there are extensively more frontal training samples compared to profile ones. In addition, it is intrinsically hard to learn a deep representation that is geometrically invariant to large pose variations. In this study, we hypothesize that there is an inherent mapping between frontal and profile faces, and consequently, their discrepancy in the deep representation space can be bridged by an equivariant mapping. To exploit this mapping, we formulate a novel Deep Residual EquivAriant Mapping (DREAM) block, which is capable of adaptively adding residuals to the input deep representation to transform a profile face representation to a canonical pose that simplifies recognition. The DREAM block consistently enhances the performance of profile face recognition for many strong deep networks, including ResNet models, without deliberately augmenting training data of profile faces. The block is easy to use, light-weight, and can be implemented with a negligible computational overhead1.

High-performance object detection relies on expensive convolutional networks to compute features, often leading to significant challenges in applications, e.g. those that require detecting objects from video streams in real time. The key to this problem is to trade accuracy for efficiency in an effective way, i.e. reducing the computing cost while maintaining competitive performance. To seek a good balance, previous efforts usually focus on optimizing the model architectures. This paper explores an alternative approach, that is, to reallocate the computation over a scale-time space. The basic idea is to perform expensive detection sparsely and propagate the results across both scales and time with substantially cheaper networks, by exploiting the strong correlations among them. Specifically, we present a unified framework that integrates detection, temporal propagation, and across-scale refinement on a Scale-Time Lattice. On this framework, one can explore various strategies to balance performance and cost. Taking advantage of this flexibility, we further develop an adaptive scheme with the detector invoked on demand and thus obtain improved tradeoff. On ImageNet VID dataset, the proposed method can achieve a competitive mAP 79.6% at 20 fps, or 79.0% at 62 fps as a performance/speed tradeoff.1

FlowNet2 [14], the state-of-the-art convolutional neural network (CNN) for optical flow estimation, requires over 160M parameters to achieve accurate flow estimation. In this paper we present an alternative network that attains performance on par with FlowNet2 on the challenging Sintel final pass and KITTI benchmarks, while being 30 times smaller in the model size and 1.36 times faster in the running speed. This is made possible by drilling down to architectural details that might have been missed in the current frameworks: (1) We present a more effective flow inference approach at each pyramid level through a lightweight cascaded network. It not only improves flow estimation accuracy through early correction, but also permits seamless incorporation of descriptor matching in our network. (2) We present a novel flow regularization layer to ameliorate the issue of outliers and vague flow boundaries by using a feature-driven local convolution. (3) Our network owns an effective structure for pyramidal feature extraction and embraces feature warping rather than image warping as practiced in FlowNet2. Our code and trained models are available at github.com/twhui/LiteFlowNet.

Despite that convolutional neural networks (CNN) have recently demonstrated high-quality reconstruction for single-image super-resolution (SR), recovering natural and realistic texture remains a challenging problem. In this paper, we show that it is possible to recover textures faithful to semantic classes. In particular, we only need to modulate features of a few intermediate layers in a single network conditioned on semantic segmentation probability maps. This is made possible through a novel Spatial Feature Transform (SFT) layer that generates affine transformation parameters for spatial-wise feature modulation. SFT layers can be trained end-to-end together with the SR network using the same loss function. During testing, it accepts an input image of arbitrary size and generates a high-resolution image with just a single forward pass conditioned on the categorical priors. Our final results show that an SR network equipped with SFT can generate more realistic and visually pleasing textures in comparison to state-of-the-art SRGAN [27] and EnhanceNet [38].

We investigate a novel approach for image restoration by reinforcement learning. Unlike existing studies that mostly train a single large network for a specialized task, we prepare a toolbox consisting of small-scale convolutional networks of different complexities and specialized in different tasks. Our method, RL-Restore, then learns a policy to select appropriate tools from the toolbox to progressively restore the quality of a corrupted image. We formulate a stepwise reward function proportional to how well the image is restored at each step to learn the action policy. We also devise a joint learning scheme to train the agent and tools for better performance in handling uncertainty. In comparison to conventional human-designed networks, RL-Restore is capable of restoring images corrupted with complex and unknown distortions in a more parameter-efficient manner using the dynamically formed toolchain1.

A cognitive robot usually needs to perform multiple tasks in practice and needs to locate the desired area for each task. Since deep learning has achieved substantial progress in image recognition, to solve this area detection problem, it is straightforward to label a functional area (affordance) image dataset and apply a well-trained deep-model-based classifier on all the potential image regions. However, annotating the functional area is time consuming and the requirement of large amount of training data limits the application scope. We observe that the functional area are usually related to the surrounding object context. In this work, we propose to use the existing object detection dataset and employ the object context as effective prior to improve the performance without additional annotated data. In particular, we formulate a two-stream network that fuses the object-related and functionality-related feature for functional area detection. The whole system is formulated in an end-to-end manner and easy to implement with current object detection framework. Experiments demonstrate that the proposed network outperforms current method by almost 20% in terms of precision and recall.

We introduce EnhanceGAN, an adversarial learning based model that performs automatic image enhancement. Traditional image enhancement frameworks typically involve training models in a fully-supervised manner, which require expensive annotations in the form of aligned image pairs. In contrast to these approaches, our proposed EnhanceGAN only requires weak supervision (binary labels on image aesthetic quality) and is able to learn enhancement operators for the task of aesthetic-based image enhancement. In particular, we show the effectiveness of a piecewise color enhancement module trained with weak supervision, and extend the proposed EnhanceGAN framework to learning a deep filtering-based aesthetic enhancer. The full differentiability of our image enhancement operators enables the training of EnhanceGAN in an end-to-end manner. We further demonstrate the capability of EnhanceGAN in learning aesthetic-based image cropping without any groundtruth cropping pairs. Our weakly-supervised EnhanceGAN reports competitive quantitative results on aesthetic-based color enhancement as well as automatic image cropping, and a user study confirms that our image enhancement results are on par with or even preferred over professional enhancement.

Many classic methods have shown non-local self-similarity in natural images to be an effective prior for image restoration. However, it remains unclear and challenging to make use of this intrinsic property via deep networks. In this paper, we propose a non-local recurrent network (NLRN) as the first attempt to incorporate non-local operations into a recurrent neural network (RNN) for image restoration. The main contributions of this work are: (1) Unlike existing methods that measure self-similarity in an isolated manner, the proposed non-local module can be flexibly integrated into existing deep networks for end-to-end training to capture deep feature correlation between each location and its neighborhood. (2) We fully employ the RNN structure for its parameter efficiency and allow deep feature correlation to be propagated along adjacent recurrent states. This new design boosts robustness against inaccurate correlation estimation due to severely degraded images. (3) We show that it is essential to maintain a confined neighborhood for computing deep feature correlation given degraded images. This is in contrast to existing practice that deploys the whole image. Extensive experiments on both image denoising and super-resolution tasks are conducted. Thanks to the recurrent non-local operations and correlation propagation, the proposed NLRN achieves superior results to state-of-the-art methods with many fewer parameters.

Semantic segmentation tasks can be well modeled by Markov Random Field (MRF). This paper addresses semantic segmentation by incorporating high-order relations and mixture of label contexts into MRF. Unlike previous works that optimized MRFs using iterative algorithm, we solve MRF by proposing a Convolutional Neural Network (CNN), namely Deep Parsing Network (DPN), which enables deterministic end-to-end computation in a single forward pass. Specifically, DPN extends a contemporary CNN to model unary terms and additional layers are devised to approximate the mean field (MF) algorithm for pairwise terms. It has several appealing properties. First, different from the recent works that required many iterations of MF during back-propagation, DPN is able to achieve high performance by approximating one iteration of MF. Second, DPN represents various types of pairwise terms, making many existing models as its special cases. Furthermore, pairwise terms in DPN provide a unified framework to encode rich contextual information in high-dimensional data, such as images and videos. Third, DPN makes MF easier to be parallelized and speeded up, thus enabling efficient inference. DPN is thoroughly evaluated on standard semantic image/video segmentation benchmarks, where a single DPN model yields state-of-the-art segmentation accuracies on PASCAL VOC 2012, Cityscapes dataset and CamVid dataset.

We propose a deep convolutional neural network (CNN) for face detection leveraging on facial attributes based supervision. We observe a phenomenon that part detectors emerge within CNN trained to classify attributes from uncropped face images, without any explicit part supervision. The observation motivates a new method for finding faces through scoring facial parts responses by their spatial structure and arrangement. The scoring mechanism is data-driven, and carefully formulated considering challenging cases where faces are only partially visible. This consideration allows our network to detect faces under severe occlusion and unconstrained pose variations. Our method achieves promising performance on popular benchmarks including FDDB, PASCAL Faces, AFW, and WIDER FACE.

The use of color in QR codes brings extra data capacity, but also inflicts tremendous challenges on the decoding process due to chromatic distortion-cross-channel color interference and illumination variation. Particularly, we further discover a new type of chromatic distortion in high-density color QR codes-cross-module color interference-caused by the high density which also makes the geometric distortion correction more challenging. To address these problems, we propose two approaches, LSVM-CMI and QDA-CMI, which jointly model these different types of chromatic distortion. Extended from SVM and QDA, respectively, both LSVM-CMI and QDA-CMI optimize over a particular objective function and learn a color classifier. Furthermore, a robust geometric transformation method and several pipeline refinements are proposed to boost the decoding performance for mobile applications. We put forth and implement a framework for high-capacity color QR codes equipped with our methods, called HiQ. To evaluate the performance of HiQ, we collect a challenging large-scale color QR code dataset, CUHK-CQRC, which consists of 5390 high-density color QR code samples. The comparison with the baseline method [2] on CUHK-CQRC shows that HiQ at least outperforms [2] by 188% in decoding success rate and 60% in bit error rate. Our implementation of HiQ in iOS and Android also demonstrates the effectiveness of our framework in real-world applications.

Data imbalance is common in many vision tasks where one or more classes are rare. Without addressing this issue, conventional methods tend to be biased toward the majority class with poor predictive accuracy for the minority class. These methods further deteriorate on small, imbalanced data that have a large degree of class overlap. In this paper, we propose a novel discriminative sparse neighbor approximation (DSNA) method to ameliorate the effect of class-imbalance during prediction. Specifically, given a test sample, we first traverse it through a cost-sensitive decision forest to collect a good subset of training examples in its local neighborhood. Then, we generate from this subset several class-discriminating but overlapping clusters and model each as an affine subspace. From these subspaces, the proposed DSNA iteratively seeks an optimal approximation of the test sample and outputs an unbiased prediction. We show that our method not only effectively mitigates the imbalance issue, but also allows the prediction to extrapolate to unseen data. The latter capability is crucial for achieving accurate prediction on small data set with limited samples. The proposed imbalanced learning method can be applied to both classification and regression tasks at a wide range of imbalance levels. It significantly outperforms the state-of-the-art methods that do not possess an imbalance handling mechanism, and is found to perform comparably or even better than recent deep learning methods by using hand-crafted features only.

Despite the remarkable progress in recent years, detecting objects in a new context remains a challenging task. Detectors learned from a public dataset can only work with a fixed list of categories, while training from scratch usually requires a large amount of training data with detailed annotations. This work aims to explore a novel approach – learning object detectors from documentary films in a weakly supervised manner. This is inspired by the observation that documentaries often provide dedicated exposition of certain object categories, where visual presentations are aligned with subtitles. We believe that object detectors can be learned from such a rich source of information. Towards this goal, we develop a joint probabilistic framework, where individual pieces of information, including video frames and subtitles, are brought together via both visual and linguistic links. On top of this formulation, we further derive a weakly supervised learning algorithm, where object model learning and training set mining are unified in an optimization procedure. Experimental results on a real world dataset demonstrate that this is an effective approach to learning new object detectors.

We propose a novel deep layer cascade (LC) method to improve the accuracy and speed of semantic segmentation. Unlike the conventional model cascade (MC) that is composed of multiple independent models, LC treats a single deep model as a cascade of several sub-models. Earlier sub-models are trained to handle easy and confident regions, and they progressively feed-forward harder regions to the next sub-model for processing. Convolutions are only calculated on these regions to reduce computations. The proposed method possesses several advantages. First, LC classifies most of the easy regions in the shallow stage and makes deeper stage focuses on a few hard regions. Such an adaptive and difficulty-aware learning improves segmentation performance. Second, LC accelerates both training and testing of deep network thanks to early decisions in the shallow stage. Third, in comparison to MC, LC is an end-to-end trainable framework, allowing joint learning of all sub-models. We evaluate our method on PASCAL VOC and Cityscapes datasets, achieving state-of-the-art performance and fast speed.

A number of studies have shown that increasing the depth or width of convolutional networks is a rewarding approach to improve the performance of image recognition. In our study, however, we observed difficulties along both directions. On one hand, the pursuit for very deep networks is met with a diminishing return and increased training difficulty, on the other hand, widening a network would result in a quadratic growth in both computational cost and memory demand. These difficulties motivate us to explore structural diversity in designing deep networks, a new dimension beyond just depth and width. Specifically, we present a new family of modules, namely the PolyInception, which can be flexibly inserted in isolation or in a composition as replacements of different parts of a network. Choosing PolyInception modules with the guidance of architectural efficiency can improve the expressive power while preserving comparable computational cost. The Very Deep PolyNet, designed following this direction, demonstrates substantial improvements over the state-of-the-art on the ILSVRC 2012 benchmark. Compared to Inception-ResNet-v2, it reduces the top-5 validation error on single crops from 4.9% to 4.25%, and that on multi-crops from 3.7% to 3.45%.

The ability to ask questions is a powerful tool to gather information in order to learn about the world and resolve ambiguities. In this paper, we explore a novel problem of generating discriminative questions to help disambiguate visual instances. Our work can be seen as a complement and new extension to the rich research studies on image captioning and question answering. We introduce the first large-scale dataset with over 10,000 carefully annotated images-question tuples to facilitate benchmarking. In particular, each tuple consists of a pair of images and 4.6 discriminative questions (as positive samples) and 5.9 non-discriminative questions (as negative samples) on average. In addition, we present an effective method for visual discriminative question generation. The method can be trained in a weakly supervised manner without discriminative images-question tuples but just existing visual question answering datasets. Promising results are shown against representative baselines through quantitative evaluations and user studies.

We present a novel and effective approach for generating new clothing on a wearer through generative adversarial learning. Given an input image of a person and a sentence describing a different outfit, our model “redresses” the person as desired, while at the same time keeping the wearer and her/his pose unchanged. Generating new outfits with precise regions conforming to a language description while retaining wearer's body structure is a new challenging task. Existing generative adversarial networks are not ideal in ensuring global coherence of structure given both the input photograph and language description as conditions. We address this challenge by decomposing the complex generative process into two conditional stages. In the first stage, we generate a plausible semantic segmentation map that obeys the wearer's pose as a latent spatial arrangement. An effective spatial constraint is formulated to guide the generation of this semantic segmentation map. In the second stage, a generative model with a newly proposed compositional mapping layer is used to render the final image with precise regions and textures conditioned on this map. We extended the DeepFashion dataset [8] by collecting sentence descriptions for 79K images. We demonstrate the effectiveness of our approach through both quantitative and qualitative evaluations. A user study is also conducted.

An active object recognition system has the advantage of acting in the environment to capture images that are more suited for training and lead to better performance at test time. In this paper, we utilize deep convolutional neural networks for active object recognition by simultaneously predicting the object label and the next action to be performed on the object with the aim of improving recognition performance. We treat active object recognition as a reinforcement learning problem and derive the cost function to train the network for joint prediction of the object label and the action. A generative model of object similarities based on the Dirichlet distribution is proposed and embedded in the network for encoding the state of the system. The training is carried out by simultaneously minimizing the label and action prediction errors using gradient descent. We empirically show that the proposed network is able to predict both the object label and the actions on GERMS, a dataset for active object recognition. We compare the test label prediction accuracy of the proposed model with Dirichlet and Naive Bayes state encoding. The results of experiments suggest that the proposed model equipped with Dirichlet state encoding is superior in performance, and selects images that lead to better training and higher accuracy of label prediction at test time.

In this paper, we propose deformable deep convolutional neural networks for generic object detection. This new deep learning object detection framework has innovations in multiple aspects. In the proposed new deep architecture, a new deformation constrained pooling (def-pooling) layer models the deformation of object parts with geometric constraint and penalty. A new pre-training strategy is proposed to learn feature representations more suitable for the object detection task and with good generalization capability. By changing the net structures, training strategies, adding and removing some key components in the detection pipeline, a set of models with large diversity are obtained, which significantly improves the effectiveness of model averaging. The proposed approach improves the mean averaged precision obtained by RCNN [16], which was the state-of-the-art, from 31% to 50.3% on the ILSVRC2014 detection test set. It also outperforms the winner of ILSVRC2014, GoogLeNet, by 6.1%. Detailed component-wise analysis is also provided through extensive experimental evaluation, which provides a global view for people to understand the deep learning object detection pipeline.

Crowd video analysis is one of the hallmark tasks of crowded scene understanding. While we observe a tremendous progress in image-based tasks with the rise of convolutional neural networks (CNNs), performance on video analysis has not (yet) attained the same level of success. In this paper, we introduce intuitive but effective temporal-aware crowd motion channels by uniformly slicing the video volume from different dimensions. Multiple CNN structures with different data-fusion strategies and weight-sharing schemes are proposed to learn the connectivity both spatially and temporally from these motion channels. To well demonstrate our deep model, we construct a new large-scale Who do What at someWhere crowd data set with 10 000 videos from 8257 crowded scenes, and build an attribute set with 94 attributes. Extensive experiments on crowd video attribute prediction demonstrate the effectiveness of our novel method over the state-of-the-art.

Groups are the primary entities that make up a crowd. Understanding group-level dynamics and properties is thus scientifically important and practically useful in a wide range of applications, especially for crowd understanding. In this paper, we show that fundamental group-level properties, such as intra-group stability and inter-group conflict, can be systematically quantified by visual descriptors. This is made possible through learning a novel collective transition prior, which leads to a robust approach for group segregation in public spaces. From the former, we further devise a rich set of group-property visual descriptors. These descriptors are scene-independent and can be effectively applied to public scenes with a variety of crowd densities and distributions. Extensive experiments on hundreds of public scene video clips demonstrate that such property descriptors are complementary to each other, scene-independent, and they convey critical information on physical states of a crowd. The proposed group-level descriptors show promising results and potentials in multiple applications, including crowd dynamic monitoring, crowd video classification, and crowd video retrieval.

We develop a Deep-Text Recurrent Network (DTRN)that regards scene text reading as a sequence labelling problem. We leverage recent advances of deep convolutional neural networks to generate an ordered highlevel sequence from a whole word image, avoiding the difficult character segmentation problem. Then a deep recurrent model, building on long short-term memory (LSTM), is developed to robustly recognize the generated CNN sequences, departing from most existing approaches recognising each character independently. Our model has a number of appealing properties in comparison to existing scene text recognition methods: (i) It can recognise highly ambiguous words by leveraging meaningful context information, allowing it to work reliably without either pre- or post-processing; (ii) the deep CNN feature is robust to various image distortions; (iii) it retains the explicit order information in word image, which is essential to discriminate word strings; (iv) the model does not depend on pre-defined dictionary, and it can process unknown words and arbitrary strings. It achieves impressive results on several benchmarks, advancing the-state-of-the-art substantially.

Data in vision domain often exhibit highly-skewed class distribution, i.e., most data belong to a few majority classes, while the minority classes only contain a scarce amount of instances. To mitigate this issue, contemporary classification methods based on deep convolutional neural network (CNN) typically follow classic strategies such as class re-sampling or cost-sensitive training. In this paper, we conduct extensive and systematic experiments to validate the effectiveness of these classic schemes for representation learning on class-imbalanced data. We further demonstrate that more discriminative deep representation can be learned by enforcing a deep network to maintain both intercluster and inter-class margins. This tighter constraint effectively reduces the class imbalance inherent in the local data neighborhood. We show that the margins can be easily deployed in standard deep learning framework through quintuplet instance sampling and the associated triple-header hinge loss. The representation learned by our approach, when combined with a simple k-nearest neighbor (kNN) algorithm, shows significant improvements over existing methods on both high-and low-level vision classification tasks that exhibit imbalanced class distribution.

Attributes offer useful mid-level features to interpret visual data. While most attribute learning methods are supervised by costly human-generated labels, we introduce a simple yet powerful unsupervised approach to learn and predict visual attributes directly from data. Given a large unlabeled image collection as input, we train deep Convolutional Neural Networks (CNNs) to output a set of discriminative, binary attributes often with semantic meanings. Specifically, we first train a CNN coupled with unsupervised discriminative clustering, and then use the cluster membership as a soft supervision to discover shared attributes from the clusters while maximizing their separability. The learned attributes are shown to be capable of encoding rich imagery properties from both natural images and contour patches. The visual representations learned in this way are also transferrable to other tasks such as object detection. We show other convincing results on the related tasks of image retrieval and classification, and contour detection.

Learning and capturing both appearance and dynamic representations are pivotal for crowd video understanding. Convolutional Neural Networks (CNNs) have shown its remarkable potential in learning appearance representations from images. However, the learning of dynamic representation, and how it can be effectively combined with appearance features for video analysis, remains an open problem. In this study, we propose a novel spatio-temporal CNN, named Slicing CNN (S-CNN), based on the decomposition of 3D feature maps into 2D spatio-and 2D temporal-slices representations. The decomposition brings unique advantages: (1) the model is capable of capturing dynamics of different semantic units such as groups and objects, (2) it learns separated appearance and dynamic representations while keeping proper interactions between them, and (3) it exploits the selectiveness of spatial filters to discard irrelevant background clutter for crowd understanding. We demonstrate the effectiveness of the proposed S-CNN model on the WWW crowd video dataset for attribute recognition and observe significant performance improvements to the state-of-the-art methods (62.55% from 51.84% [21]).

Face detection is one of the most studied topics in the computer vision community. Much of the progresses have been made by the availability of face detection benchmark datasets. We show that there is a gap between current face detection performance and the real world requirements. To facilitate future face detection research, we introduce the WIDER FACE dataset1, which is 10 times larger than existing datasets. The dataset contains rich annotations, including occlusions, poses, event categories, and face bounding boxes. Faces in the proposed dataset are extremely challenging due to large variations in scale, pose and occlusion, as shown in Fig. 1. Furthermore, we show that WIDER FACE dataset is an effective training source for face detection. We benchmark several representative detection systems, providing an overview of state-of-the-art performance and propose a solution to deal with large scale variation. Finally, we discuss common failure cases that worth to be further investigated.

We investigate the problem of fine-grained sketch-based image retrieval (SBIR), where free-hand human sketches are used as queries to perform instance-level retrieval of images. This is an extremely challenging task because (i) visual comparisons not only need to be fine-grained but also executed cross-domain, (ii) free-hand (finger) sketches are highly abstract, making fine-grained matching harder, and most importantly (iii) annotated cross-domain sketch-photo datasets required for training are scarce, challenging many state-of-the-art machine learning techniques. In this paper, for the first time, we address all these challenges, providing a step towards the capabilities that would underpin a commercial sketch-based image retrieval application. We introduce a new database of 1,432 sketchphoto pairs from two categories with 32,000 fine-grained triplet ranking annotations. We then develop a deep tripletranking model for instance-level SBIR with a novel data augmentation and staged pre-training strategy to alleviate the issue of insufficient fine-grained training data. Extensive experiments are carried out to contribute a variety of insights into the challenges of data sufficiency and over-fitting avoidance when training deep networks for finegrained cross-domain ranking tasks.

We present a practical approach to address the problem of unconstrained face alignment for a single image. In our unconstrained problem, we need to deal with large shape and appearance variations under extreme head poses and rich shape deformation. To equip cascaded regressors with the capability to handle global shape variation and irregular appearance-shape relation in the unconstrained scenario, we partition the optimisation space into multiple domains of homogeneous descent, and predict a shape as a composition of estimations from multiple domain-specific regressors. With a specially formulated learning objective and a novel tree splitting function, our approach is capable of estimating a robust and meaningful composition. In addition to achieving state-of-the-art accuracy over existing approaches, our framework is also an efficient solution (350 FPS), thanks to the on-the-fly domain exclusion mechanism and the capability of leveraging the fast pixel feature.

Color brings extra data capacity for QR codes, but it also brings tremendous challenges to the decoding because of color interference and illumination variation, especially for high-density QR codes. In this paper, we put forth a framework for high-capacity QR codes, HiQ, which optimizes the decoding algorithm for high-density QR codes to achieve robust and fast decoding on mobile devices, and adopts a learning-based approach for color recovery. Moreover, we propose a robust geometric transformation algorithm to correct the geometric distortion. We also provide a challenging color QR code dataset, CUHK-CQRC, which consists of 5390 high-density color QR code samples captured by different smartphones under different lighting conditions. Experimental results show that HiQ outperforms the baseline [1] by 286% in decoding success rate and 60% in bit error rate.

Existing deep embedding methods in vision tasks are capable of learning a compact Euclidean space from images, where Euclidean distances correspond to a similarity metric. To make learning more effective and efficient, hard sample mining is usually employed, with samples identified through computing the Euclidean feature distance. However, the global Euclidean distance cannot faithfully characterize the true feature similarity in a complex visual feature space, where the intraclass distance in a high-density region may be larger than the interclass distance in low-density regions. In this paper, we introduce a Position-Dependent Deep Metric (PDDM) unit, which is capable of learning a similarity metric adaptive to local feature structure. The metric can be used to select genuinely hard samples in a local neighborhood to guide the deep embedding learning in an online and robust manner. The new layer is appealing in that it is pluggable to any convolutional networks and is trained end-to-end. Our local similarity-aware feature embedding not only demonstrates faster convergence and boosted performance on two complex image retrieval datasets, its large margin nature also leads to superior generalization results under the large and open set scenarios of transfer learning and zero-shot learning on ImageNet 2010 and ImageNet-10K datasets.

We propose a deep learning method for single image super-resolution (SR). Our method directly learns an end-to-end mapping between the low/high-resolution images. The mapping is represented as a deep convolutional neural network (CNN) that takes the low-resolution image as the input and outputs the high-resolution one. We further show that traditional sparse-coding-based SR methods can also be viewed as a deep convolutional network. But unlike traditional methods that handle each component separately, our method jointly optimizes all layers. Our deep CNN has a lightweight structure, yet demonstrates state-of-the-art restoration quality, and achieves fast speed for practical on-line usage. We explore different network structures and parameter settings to achieve trade-offs between performance and speed. Moreover, we extend our network to cope with three color channels simultaneously, and show better overall reconstruction quality.

In this study, we show that landmark detection or face alignment task is not a single and independent problem. Instead, its robustness can be greatly improved with auxiliary information. Specifically, we jointly optimize landmark detection together with the recognition of heterogeneous but subtly correlated facial attributes, such as gender, expression, and appearance attributes. This is non-trivial since different attribute inference tasks have different learning difficulties and convergence rates. To address this problem, we formulate a novel tasks-constrained deep model, which not only learns the inter-task correlation but also employs dynamic task coefficients to facilitate the optimization convergence when learning multiple complex tasks. Extensive evaluations show that the proposed task-constrained learning (i) outperforms existing face alignment methods, especially in dealing with faces with severe occlusion and pose variation, and (ii) reduces model complexity drastically compared to the state-of-the-art methods based on cascaded deep model.

While clustering is usually an unsupervised operation, there are circumstances where we have access to prior belief that pairs of samples should (or should not) be assigned with the same cluster. Constrained clustering aims to exploit this prior belief as constraint (or weak supervision) to influence the cluster formation so as to obtain a data structure more closely resembling human perception. Two important issues remain open: 1) how to exploit sparse constraints effectively and 2) how to handle ill-conditioned/noisy constraints generated by imperfect oracles. In this paper, we present a novel pairwise similarity measure framework to address the above issues. Specifically, in contrast to existing constrained clustering approaches that blindly rely on all features for constraint propagation, our approach searches for neighborhoods driven by discriminative feature selection for more effective constraint diffusion. Crucially, we formulate a novel approach to handling the noisy constraint problem, which has been unrealistically ignored in the constrained clustering literature. Extensive comparative results show that our method is superior to the state-of-the-art constrained clustering approaches and can generally benefit existing pairwise similarity-based data clustering algorithms, such as spectral clustering and affinity propagation.

We consider the problem of learning deep representation when target labels are available. In this paper, we show that there exists intrinsic relationship between target coding and feature representation learning in deep networks. Specifically, we found that distributed binary acode with error correcting capability is more capable of encouraging discriminative features, in comparison tothe 1-of-K coding that is typically used in supervised deep learning. This new finding reveals additional benefit of using error-correcting code for deep model learning,apart from its well-known error correcting property. Extensive experiments are conducted on popular visual benchmark datasets.

In this paper, we propose deformable deep convolutional neural networks for generic object detection. This new deep learning object detection framework has innovations in multiple aspects. In the proposed new deep architecture, a new deformation constrained pooling (def-pooling) layer models the deformation of object parts with geometric constraint and penalty. A new pre-training strategy is proposed to learn feature representations more suitable for the object detection task and with good generalization capability. By changing the net structures, training strategies, adding and removing some key components in the detection pipeline, a set of models with large diversity are obtained, which significantly improves the effectiveness of model averaging. The proposed approach improves the mean averaged precision obtained by RCNN [14], which was the state-of-the-art, from 31% to 50.3% on the ILSVRC2014 detection test set. It also outperforms the winner of ILSVRC2014, GoogLeNet, by 6.1%. Detailed component-wise analysis is also provided through extensive experimental evaluation, which provide a global view for people to understand the deep learning object detection pipeline.

Crowded scene understanding is a fundamental problem in computer vision. In this study, we develop a multi-task deep model to jointly learn and combine appearance and motion features for crowd understanding. We propose crowd motion channels as the input of the deep model and the channel design is inspired by generic properties of crowd systems. To well demonstrate our deep model, we construct a new large-scale WWW Crowd dataset with 10, 000 videos from 8, 257 crowded scenes, and build an attribute set with 94 attributes on WWW. We further measure user study performance on WWW and compare this with the proposed deep models. Extensive experiments show that our deep models display significant performance improvements in cross-scene attribute recognition compared to strong crowd-related feature-based baselines, and the deeply learned features behave a superior performance in multi-task learning.

This paper aims to highlight vision related tasks centered around “car”, which has been largely neglected by vision community in comparison to other objects. We show that there are still many interesting car-related problems and applications, which are not yet well explored and researched. To facilitate future car-related research, in this paper we present our on-going effort in collecting a large-scale dataset, “CompCars”, that covers not only different car views, but also their different internal and external parts, and rich attributes. Importantly, the dataset is constructed with a cross-modality nature, containing a surveillance-nature set and a web-nature set. We further demonstrate a few important applications exploiting the dataset, namely car model classification, car model verification, and attribute prediction. We also discuss specific challenges of the car-related problems and other potential applications that worth further investigations. The latest dataset can be downloaded at http://mmlab.ie.cuhk.edu.hk/ datasets/comp_cars/index.html.

We present a novel face alignment framework based on coarse-to-fine shape searching. Unlike the conventional cascaded regression approaches that start with an initial shape and refine the shape in a cascaded manner, our approach begins with a coarse search over a shape space that contains diverse shapes, and employs the coarse solution to constrain subsequent finer search of shapes. The unique stage-by-stage progressive and adaptive search i) prevents the final solution from being trapped in local optima due to poor initialisation, a common problem encountered by cascaded regression approaches; and ii) improves the robustness in coping with large pose variations. The framework demonstrates real-time performance and state-of-the-art results on various benchmarks including the challenging 300-W dataset.

Lossy compression introduces complex compression artifacts, particularly the blocking artifacts, ringing effects and blurring. Existing algorithms either focus on removing blocking artifacts and produce blurred output, or restores sharpened images that are accompanied with ringing effects. Inspired by the deep convolutional networks (DCN) on super-resolution, we formulate a compact and efficient network for seamless attenuation of different compression artifacts. We also demonstrate that a deeper model can be effectively trained with the features learned in a shallow network. Following a similar "easy to hard" idea, we systematically investigate several practical transfer settings and show the effectiveness of transfer learning in low level vision problems. Our method shows superior performance than the state-of-the-arts both on the benchmark datasets and the real-world use cases (i.e. Twitter).

This paper addresses semantic image segmentation by incorporating rich information into Markov Random Field (MRF), including high-order relations and mixture of label contexts. Unlike previous works that optimized MRFs using iterative algorithm, we solve MRF by proposing a Convolutional Neural Network (CNN), namely Deep Parsing Network (DPN), which enables deterministic end-to-end computation in a single forward pass. Specifically, DPN extends a contemporary CNN architecture to model unary terms and additional layers are carefully devised to approximate the mean field algorithm (MF) for pairwise terms. It has several appealing properties. First, different from the recent works that combined CNN and MRF, where many iterations of MF were required for each training image during back-propagation, DPN is able to achieve high performance by approximating one iteration of MF. Second, DPN represents various types of pairwise terms, making many existing works as its special cases. Third, DPN makes MF easier to be parallelized and speeded up in Graphical Processing Unit (GPU). DPN is thoroughly evaluated on the PASCAL VOC 2012 dataset, where a single DPN model yields a new state-of-the-art segmentation accuracy of 77.5%.

In this paper, we propose a novel deep convolutional network (DCN) that achieves outstanding performance on FDDB, PASCAL Face, and AFW. Specifically, our method achieves a high recall rate of 90.99% on the challenging FDDB benchmark, outperforming the state-of-the-art method [23] by a large margin of 2.91%. Importantly, we consider finding faces from a new perspective through scoring facial parts responses by their spatial structure and arrangement. The scoring mechanism is carefully formulated considering challenging cases where faces are only partially visible. This consideration allows our network to detect faces under severe occlusion and unconstrained pose variation, which are the main difficulty and bottleneck of most existing face detection approaches. We show that despite the use of DCN, our network can achieve practical runtime speed.

Social relation defines the association, e.g., warm, friendliness, and dominance, between two or more people. Motivated by psychological studies, we investigate if such fine grained and high-level relation traits can be characterised and quantified from face images in the wild. To address this challenging problem we propose a deep model that learns a rich face representation to capture gender, expression, head pose, and age-related attributes, and then performs pairwise-face reasoning for relation prediction. To learn from heterogeneous attribute sources, we formulate a new network architecture with a bridging layer to leverage the inherent correspondences among these datasets. It can also cope with missing target attribute labels. Extensive experiments show that our approach is effective for fine-grained social relation learning in images and videos.

Groups are the primary entities that make up a crowd. Understanding group-level dynamics and properties is thus scientifically important and practically useful in a wide range of applications, especially for crowd understanding. In this study we show that fundamental group-level properties, such as intra-group stability and inter-group conflict, can be systematically quantified by visual descriptors. This is made possible through learning a novel Collective Transition prior, which leads to a robust approach for group segregation in public spaces. From the prior, we further devise a rich set of group property visual descriptors. These descriptors are scene-independent, and can be effectively applied to public-scene with variety of crowd densities and distributions. Extensive experiments on hundreds of public scene video clips demonstrate that such property descriptors are not only useful but also necessary for group state analysis and crowd scene understanding.

Spectral clustering requires robust and meaningful affinity graphs as input in order to form clusters with desired structures that can well support human intuition. To construct such affinity graphs is non-trivial due to the ambiguity and uncertainty inherent in the raw data. In contrast to most existing clustering methods that typically employ all available features to construct affinity matrices with the Euclidean distance, which is often not an accurate representation of the underlying data structures, we propose a novel unsupervised approach to generating more robust affinity graphs via identifying and exploiting discriminative features for improving spectral clustering. Specifically, our model is capable of capturing and combining subtle similarity information distributed over discriminative feature subspaces for more accurately revealing the latent data distribution and thereby leading to improved data clustering, especially with heterogeneous data sources. We demonstrate the efficacy of the proposed approach on challenging image and video datasets.

It is common for CCTV operators to overlook interesting events taking place within the crowd due to large number of people in the crowded scene (i.e. marathon, rally). Thus, there is a dire need to automate the detection of salient crowd regions acquiring immediate attention for a more effective and proactive surveillance. This paper proposes a novel framework to identify and localize salient regions in a crowd scene, by transforming low-level features extracted from crowd motion field into a global similarity structure. The global similarity structure representation allows the discovery of the intrinsic manifold of the motion dynamics, which could not be captured by the low-level representation. Ranking is then performed on the global similarity structure to identify a set of extrem a. The proposed approach is unsupervised so learning stage is eliminated. Experimental results on public datasets demonstrates the effectiveness of exploiting such extrem a in identifying salient regions in various crowd scenarios that exhibit crowding, local irregular motion, and unique motion areas such as sources and sinks.

The capability of recognizing pedestrian attributes, such as gender and clothing style, at far distance, is of practical interest in far-view surveillance scenarios where face and body close-shots are hardly available. We make two contributions in this paper. First, we release a new pedestrian attribute dataset, which is by far the largest and most diverse of its kind. We show that the large-scale dataset facilitates the learning of robust attribute detectors with good generalization performance. Second, we present the benchmark performance by SVM-based method and propose an alternative approach that exploits context of neighboring pedestrian images for improved attribute inference.

A number of computer vision problems such as human age estimation, crowd density estimation and body/face pose (view angle) estimation can be formulated as a regression problem by learning a mapping function between a high dimensional vector-formed feature input and a scalar-valued output. Such a learning problem is made difficult due to sparse and imbalanced training data and large feature variations caused by both uncertain viewing conditions and intrinsic ambiguities between observable visual features and the scalar values to be estimated. Encouraged by the recent success in using attributes for solving classification problems with sparse training data, this paper introduces a novel cumulative attribute concept for learning a regression model when only sparse and imbalanced data are available. More precisely, low-level visual features extracted from sparse and imbalanced image samples are mapped onto a cumulative attribute space where each dimension has clearly defined semantic interpretation (a label) that captures how the scalar output value (e.g. age, people count) changes continuously and cumulatively. Extensive experiments show that our cumulative attribute framework gains notable advantage on accuracy for both age estimation and crowd counting when compared against conventional regression models, especially when the labelled training data is sparse with imbalanced sampling.

Owing to visual ambiguities and disparities, person re-identification methods inevitably produce sub optimal rank-list, which still requires exhaustive human eyeballing to identify the correct target from hundreds of different likely-candidates. Existing re-identification studies focus on improving the ranking performance, but rarely look into the critical problem of optimising the time-consuming and error-prone post-rank visual search at the user end. In this study, we present a novel one-shot Post-rank Optimization (POP) method, which allows a user to quickly refine their search by either "one-shot" or a couple of sparse negative selections during a re-identification process. We conduct systematic behavioural studies to understand user's searching behaviour and show that the proposed method allows correct re-identification to converge 2.6 times faster than the conventional exhaustive search. Importantly, through extensive evaluations we demonstrate that the method is capable of achieving significant improvement over the state-of-the-art distance metric learning based ranking models, even with just "one shot" feedback optimisation, by as much as over 30% performance improvement for rank 1 re-identification on the VIPeR and i-LIDS datasets.

Regression-based techniques have shown promising results for people counting in crowded scenes. However, most existing techniques require expensive and laborious data annotation for model training. In this study, we propose to address this problem from three perspectives: (1) Instead of exhaustively annotating every single frame, the most informative frames are selected for annotation automatically and actively. (2) Rather than learning from only labelled data, the abundant unlabelled data are exploited. (3) Labelled data from other scenes are employed to further alleviate the burden for data annotation. All three ideas are implemented in a unified active and semi-supervised regression framework with ability to perform transfer learning, by exploiting the underlying geometric structure of crowd patterns via manifold analysis. Extensive experiments validate the effectiveness of our approach.

Generating coherent synopsis for surveillance video stream remains a formidable challenge due to the ambiguity and uncertainty inherent to visual observations. In contrast to existing video synopsis approaches that rely on visual cues alone, we propose a novel multi-source synopsis framework capable of correlating visual data and independent non-visual auxiliary information to better describe and summarise subtle physical events in complex scenes. Specifically, our unsupervised framework is capable of seamlessly uncovering latent correlations among heterogeneous types of data sources, despite the non-trivial heteroscedasticity and dimensionality discrepancy problems. Additionally, the proposed model is robust to partial or missing non-visual information. We demonstrate the effectiveness of our framework on two crowded public surveillance datasets.

While spectral clustering is usually an unsupervised operation, there are circumstances in which we have prior belief that pairs of samples should (or should not) be assigned with the same cluster. Constrained spectral clustering aims to exploit this prior belief as constraint (or weak supervision) to influence the cluster formation so as to obtain a structure more closely resembling human perception. Two important issues remain open: (1) how to propagate sparse constraints effectively, (2) how to handle ill-conditioned/noisy constraints generated by imperfect oracles. In this paper we present a unified framework to address the above issues. Specifically, in contrast to existing constrained spectral clustering approaches that blindly rely on all features for constructing the spectral, our approach searches for neighbours driven by discriminative feature selection for more effective constraint diffusion. Crucially, we formulate a novel data-driven filtering approach to handle the noisy constraint problem, which has been unrealistically ignored in constrained spectral clustering literature.

Existing person re-identification methods conventionally rely on labelled pairwise data to learn a task-specific distance metric for ranking. The value of unlabelled gallery instances is generally overlooked. In this study, we show that it is possible to propagate the query information along the unlabelled data manifold in an unsupervised way to obtain robust ranking results. In addition, we demonstrate that the performance of existing supervised metric learning methods can be significantly boosted once integrated into the proposed manifold ranking-based framework. Extensive evaluation is conducted on three benchmark datasets.

This paper presents a multi-output regression model for crowd counting in public scenes. Existing counting by regression methods either learn a single model for global counting, or train a large number of separate regressors for localised density estimation. In contrast, our single regression model based approach is able to estimate people count in spatially localised regions and is more scalable without the need for training a large number of regressors proportional to the number of local regions. In particular, the proposed model automatically learns the functional mapping between interdependent low-level features and multi-dimensional structured outputs. The model is able to discover the inherent importance of different features for people counting at different spatial locations. Extensive evaluations on an existing crowd analysis benchmark dataset and a new more challenging dataset demonstrate the effectiveness of our approach. 1

Problem: This work systematically investigates the effectiveness of various visual feature coding schemes for facilitating the learning of timedelayed dependencies among disjoint multi-camera views. Related work: Quite a few studies [3, 4, 6] have been proposed to model inter-camera dependency across non-overlapping camera views. Learning time-delayed correlations among disjoint cameras in crowded public scenarios is a non-trivial task: (1) the time gaps between camera views are unknown therefore activities in two related views may occur at arbitrary time delays with high uncertainty; (2) the features are inevitably noisy, ambiguous, and may vary drastically across views owning to illumination condition, camera angles, and changes in object pose. Most state-ofthe-art methods typically hand pick a few features tailored to the target environment, with the hope that those chosen features contain robust and sufficient statistics for correlating the time-delayed activity patterns across disjoint views. These manual approaches to hard selection of features are neither principled nor generalisable to different scene context. Our solution: In this study, we wish to examine the concept that visual features should be coded and selected automatically for robust and accurate time-delayed dependency learning. The contributions of this study are two-fold: (1) We present a systematic study and evaluation to investigate the effectiveness of supervised and unsupervised feature coding methods to facilitate the learning of inter-camera activity pattern dependencies. (2) We systematically evaluate the sensitivity of inter-camera time delayed dependency learning given different training video sizes and region decomposition qualities. These factors are critical for accurate dependency learning but have been largely ignored by the published existing work in the literature. Approach overview: We employ the Random Forest [2] as the supervised feature coding approach. In particular, given a set of localised features extracted from a region, together with people count training label over time, we first train a regression forest to learn the non-linear mapping between the crowd density and the corresponding low-level features. Given unseen data, we then construct a time series based on the predicted crowd density ŷ obtained from the regression forest (RF pred), the treestructured code (tree code) [5], or the combination of the two. As for unsupervised coding scheme, we use the Latent Dirichlet Allocation (LDA) [1] to map the low-level features into codewords that capture the topic distribution, whereby an image region patch (document) d is treated as a collection of j = 1 . . .Ni features (words). To form the unsupervised feature codes, given a sequence of localised feature vectors detected from a region, we first perform quantisation on each feature to generate a bag-of-word representation for all image patches. Similar to text documents, these bag-of-word represented image patches are fed into the LDA, which gives us a topic-based representation. Once having the topic-based code (topic code), we perform k-means quantisation on them, producing the final compact topic-based code, and concatenate them over time to form a time series. To solve the problem of using the feature codes for learning intercamera dependencies, we adopt the Time Delayed Mutual Information (TDMI) proposed in [3] due to its reported effectiveness and simplicity. The input to TDMI are time series generated from either the supervised or the unsupervised coding scheme. In addition to measuring deviation error in transition time, we propose a new metrics to evaluate the effectiveness of different coding methods, called Mutual Information Margin (MIM):

Learning from streams of evolving and unbounded data is an important problem, for example in visual surveillance or internet scale data. For such large and evolving real-world data, exhaustive supervision is impractical, particularly so when the full space of classes is not known in advance therefore joint class discovery (exploration) and boundary learning (exploitation) becomes critical. Active learning has shown promise in jointly optimising exploration-exploitation with minimal human supervision. However, existing active learning methods either rely on heuristic multi-criteria weighting or are limited to batch processing. In this paper, we present a new unified framework for joint exploration-exploitation active learning in streams without any heuristic weighting. Extensive evaluation on classification of various image and surveillance video datasets demonstrates the superiority of our framework over existing methods.

Activity modeling and unusual event detection in a network of cameras is challenging, particularly when the camera views are not overlapped. We show that it is possible to detect unusual events in multiple disjoint cameras as context-incoherent patterns through incremental learning of time delayed dependencies between distributed local activities observed within and across camera views. Specifically, we model multicamera activities using a Time Delayed Probabilistic Graphical Model (TD-PGM) with different nodes representing activities in different decomposed regions from different views and the directed links between nodes encoding their time delayed dependencies. To deal with visual context changes, we formulate a novel incremental learning method for modeling time delayed dependencies that change over time. We validate the effectiveness of the proposed approach using a synthetic data set and videos captured from a camera network installed at a busy underground station.

We present a new approach for activity modelling and anomaly detection based on non-parametric Gaussian Process (GP) models. Specifically, GP regression models are formulated to learn non-linear relationships between multi-object activity patterns observed from semantically decomposed regions in complex scenes. Predictive distributions are inferred from the regression models to compare with the actual observations for real-time anomaly detection. The use of a flexible, non-parametric model alleviates the difficult problem of selecting appropriate model complexity encountered in parametric models such as Dynamic Bayesian Networks (DBNs). Crucially, our GP models need fewer parameters; they are thus less likely to overfit given sparse data. In addition, our approach is robust to the inevitable noise in activity representation as noise is modelled explicitly in the GP models. Experimental results on a public traffic scene show that our models outperform DBNs in terms of anomaly sensitivity, noise robustness, and flexibility in modelling complex activity.

We propose a novel approach for modelling correlations between activities in a busy public space captured by multiple non-overlapping and uncalibrated cameras. In our approach, each camera view is automatically decomposed into semantic regions, across which different spatio-temporal activity patterns are observed. A novel Cross Canonical Correlation Analysis (xCCA) framework is formulated to detect and quantify temporal and causal relationships between regional activities within and across camera views. The approach accomplishes three tasks: (1) estimate the spatial and temporal topology of the camera network; (2) facilitate more robust and accurate person re-identification; (3) perform global activity modelling and video temporal segmentation by linking visual evidence collected across camera views. Our approach differs from the state of the art in that it does not rely on either intra or inter camera tracking. It therefore can be applied to even the most challenging video surveillance settings featured with severe occlusions and extremely low spatial and temporal resolutions. Its effectiveness is demonstrated using 153 hours of videos from 8 cameras installed in a busy underground station.

We present a novel approach for detecting global behaviour anomalies in multiple disjoint cameras by learning time delayed dependencies between activities cross camera views. Specifically, we propose to model multi-camera activities using a Time Delayed Probabilistic Graphical Model (TD-PGM) with different nodes representing activities in different semantically decomposed regions from different camera views, and the directed links between nodes encoding causal relationships between the activities. A novel two-stage structure learning algorithm is formulated to learn globally optimised time-delayed dependencies. A new cumulative abnormality score is also introduced to replace the conventional log-likelihood score for gaining significantly more robust and reliable real-time anomaly detection. The effectiveness of the proposed approach is validated using a camera network installed at a busy underground station.

This paper presents a fuzzy ARTMAP (FAM) based modular architecture for multi-class pattern recognition known as modular adaptive resonance theory map (MARTMAP). The prediction of class membership is made collectively by combining outputs from multiple novelty detectors. Distance-based familiarity discrimination is introduced to improve the robustness of MARTMAP in the presence of noise. The effectiveness of the proposed architecture is analyzed and compared with ARTMAP-FD network, FAM network, and One-Against-One Support Vector Machine (OAO-SVM). Experimental results show that MARTMAP is able to retain effective familiarity discrimination in noisy environment, and yet less sensitive to class imbalance problem as compared to its counterparts.