The traditional paradigm to update retrieval models requires re-computing the embeddings of the gallery data, a time-consuming and computationally intensive process known as backfilling. To circumvent backfilling, Backward-Compatible Learning (BCL) has been widely explored, which aims to train a new model compatible with the old one. Many previous works focus on effectively aligning the embeddings of the new model with those of the old one to enhance backward compatibility. Nevertheless, such strong alignment constraints would compromise the discriminative ability of the new model, particularly when different classes are closely clustered and hard to distinguish in the old feature space. To address this issue, we propose to relax the constraints by introducing perturbations to the old feature prototypes. This allows us to align the new feature space with a pseudo-old feature space defined by these perturbed prototypes, thereby preserving the discriminative ability of the new model in backward-compatible learning. We have developed two approaches for calculating the perturbations: Neighbor-Driven Prototype Perturbation (NDPP) and Optimization-Driven Prototype Perturbation (ODPP). Particularly, they take into account the feature distributions of not only the old but also the new models to obtain proper perturbations along with new model updating. Extensive experiments on the landmark and commodity datasets demonstrate that our approaches perform favorably against state-of-the-art BCL algorithms.

Scene text editing aims to modify text content within scene images while maintaining style consistency. Traditional methods achieve this by explicitly disentangling style and content from the source image and then fusing the style with the target content, while ensuring content consistency using a pre-trained recognition model. Despite notable progress, these methods suffer from complex pipelines, leading to suboptimal performance in complex scenarios. In this work, we introduce Recognition-Synergistic Scene Text Editing (RS-STE), a novel approach that fully exploits the intrinsic synergy of text recognition for editing. Our model seamlessly integrates text recognition with text editing within a unified framework, and leverages the recognition model’s ability to implicitly disentangle style and content while ensuring content consistency. Specifically, our approach employs a multi-modal parallel decoder based on transformer architecture, which predicts both text content and stylized images in parallel. Additionally, our cyclic self-supervised fine-tuning strategy enables effective training on unpaired real-world data without ground truth, enhancing style and content consistency through a twice-cyclic generation process. Built on a relatively simple architecture, RS-STE achieves state-of-the-art performance on both synthetic and real-world benchmarks, and further demonstrates the effectiveness of leveraging the generated hard cases to boost the performance of downstream recognition tasks. Code is available at https://github.com/ZhengyaoFang/RS-STE.

Asymmetric retrieval is a typical scenario in real-world retrieval systems, where compatible models of varying capacities are deployed on platforms with different resource configurations. Existing methods generally train pre-defined networks or subnetworks with capacities specifically designed for pre-determined platforms, using compatible learning. Nevertheless, these methods suffer from limited flexibility for multi-platform deployment. For example, when introducing a new platform into the retrieval systems, developers have to train an additional model at an appropriate capacity that is compatible with existing models via backward-compatible learning. In this paper, we propose a Prunable Network with self-compatibility, which allows developers to generate compatible subnetworks at any desired capacity through post-training pruning. Thus it allows the creation of a sparse subnetwork matching the resources of the new platform without additional training. Specifically, we optimize both the architecture and weight of subnetworks at different capacities within a dense network in compatible learning. We also design a conflict-aware gradient integration scheme to handle the gradient conflicts between the dense network and subnetworks during compatible learning. Extensive experiments on diverse benchmarks and visual backbones demonstrate the effectiveness of our method. The code will be made publicly available.

Existing text-to-video methods struggle to transfer motion smoothly from a reference object to a target object with significant differences in appearance or structure between them. To address this challenge, we introduce MotionShot, a training-free framework capable of parsing reference-target correspondences in a fine-grained manner, thereby achieving high-fidelity motion transfer while preserving coherence in appearance. To be specific, MotionShot first performs semantic feature matching to ensure high-level alignments between the reference and target objects. It then further establishes low-level morphological alignments through reference-to-target shape retargeting. By encoding motion with temporal attention, our MotionShot can coherently transfer motion across objects, even in the presence of significant appearance and structure disparities, demonstrated by extensive experiments. The project page is available at: https://motionshot.github.io/.

Recent advances in point cloud perception have demonstrated remarkable progress in scene understanding through vision-language alignment leveraging large language models (LLMs). However, existing methods may still encounter challenges in handling complex instructions that require accurate spatial reasoning, even if the 3D point cloud data provides detailed spatial cues such as size and position for identifying the targets. To tackle this issue, we propose Relevant Reasoning Segmentation (R$^2$S), a reasoning-based segmentation framework. The framework emulates human cognitive processes by decomposing spatial reasoning into two sequential stages: first identifying relevant elements, then processing instructions guided by their associated visual priors. Furthermore, acknowledging the inadequacy of existing datasets in complex reasoning tasks, we introduce 3D ReasonSeg, a reasoning-based segmentation dataset comprising 25,185 training samples and 3,966 validation samples with precise annotations. Both quantitative and qualitative experiments demonstrate that the R$^2$S and 3D ReasonSeg effectively endow 3D point cloud perception with stronger spatial reasoning capabilities, and we hope that they can serve as a new baseline and benchmark for future work.

Adapting pretrained diffusion-based generative models for text-driven image editing with negligible tuning overhead has demonstrated remarkable potential. A classical adaptation paradigm, as followed by these methods, first infers the generative trajectory inversely for a given source image by image inversion, then performs image editing along the inferred trajectory guided by the target text prompts. However, the performance of image editing is heavily limited by the approximation errors introduced during image inversion by diffusion models, which arise from the absence of exact supervision in the intermediate generative steps. To circumvent this issue, we investigate the parameter-efficient adaptation of binary-quantized generative models for image editing, and leverage their inherent characteristic that the exact intermediate quantized representations of a source image are attainable, enabling more effective supervision for precise image inversion. Specifically, we propose EditInfinity, which adapts Infinity, a binary-quantized generative model, for image editing. We propose an efficient yet effective image inversion mechanism that integrates text prompting rectification and image style preservation, enabling precise image inversion. Furthermore, we devise a holistic smoothing strategy which allows our EditInfinity to perform image editing with high fidelity to source images and precise semantic alignment to the text prompts. Extensive experiments on the PIE-Bench benchmark across add, change, and delete editing operations, demonstrate the superior performance of our model compared to state-of-the-art diffusion-based baselines. Code available at: https://github.com/yx-chen-ust/EditInfinity.

Few-shot classification aims to adapt classifiers trained on base classes to novel classes with a few shots. However, the limited amount of training data is often inadequate to represent the intraclass variations in novel classes. This can result in biased estimation of the feature distribution, which in turn results in inaccurate decision boundaries, especially when the support data are outliers. To address this issue, we propose a feature enhancement method called CORrelation-guided feature Enrichment that generates improved features for novel classes using weak supervision from the base classes. The proposed CORrelation-guided feature Enhancement (CORE) method utilizes an autoencoder (AE) architecture but incorporates classification information into its latent space. This design allows the CORE to generate more discriminative features while discarding irrelevant content information. After being trained on base classes, CORE's generative ability can be transferred to novel classes that are similar to those in the base classes. By using these generative features, we can reduce the estimation bias of the class distribution, which makes few-shot learning (FSL) less sensitive to the selection of support data. Our method is generic and flexible and can be used with any feature extractor and classifier. It can be easily integrated into existing FSL approaches. Experiments with different backbones and classifiers show that our proposed method consistently outperforms existing methods on various widely used benchmarks.

This article aims to solve the video object segmentation (VOS) task in a scribble-supervised manner, in which VOS models are not only initialized with sparse target scribbles for inference but also trained by sparse scribble annotations. Thus, the annotation burdens for both initialization and training can be substantially lightened. The difficulties of scribble-supervised VOS lie in two aspects: 1) it demands a strong reasoning ability to carefully segment the target given only a sparse initial target scribble and 2) it necessitates learning dense prediction from sparse scribble annotations during training, requiring powerful learning capability. In this work, we propose a reliability-guided hierarchical memory network (RHMNet) for this task, which segments the target in a stepwise expanding strategy w.r.t. the memory reliability level. To be specific, RHMNet maintains a reliability-guided memory bank. It first uses the high-reliability memory to locate the region with high reliability belonging to the target, i.e., highly similar to the initial target scribble. Then, it expands the located high-reliability region to the entire target conditioned on the region itself and all existing memories. In addition, we propose a scribble-supervised learning mechanism to facilitate the model learning for dense prediction. It exploits the pixel-level relations within a single frame and the instance-level variations across multiple frames to take full advantage of the scribble annotations in sequence training samples. The favorable performance on four popular benchmarks demonstrates that our method is promising. Our project is available at: https://github.com/mkg1204/RHMNet-for-SSVOS.

As a prominent parameter-efficient fine-tuning technique in NLP, prompt tuning is being explored its potential in computer vision. Typical methods for visual prompt tuning follow the sequential modeling paradigm stemming from NLP, which represents an input image as a flattened sequence of token embeddings and then learns a set of unordered parameterized tokens prefixed to the sequence representation as the visual prompts for task adaptation of large vision models. While such sequential modeling paradigm of visual prompt has shown great promise, there are two potential limitations. First, the learned visual prompts cannot model the underlying spatial relations in the input image, which is crucial for image encoding. Second, since all prompt tokens play the same role of prompting for all image tokens without distinction, it lacks the fine-grained prompting capability, i.e., individual prompting for different image tokens. In this work, we propose the Spatially Aligned-and-Adapted Visual Prompt model (SA^2VP), which learns a two-dimensional prompt token map with equal (or scaled) size to the image token map, thereby being able to spatially align with the image map. Each prompt token is designated to prompt knowledge only for the spatially corresponding image tokens. As a result, our model can conduct individual prompting for different image tokens in a fine-grained manner. Moreover, benefiting from the capability of preserving the spatial structure by the learned prompt token map, our SA^2VP is able to model the spatial relations in the input image, leading to more effective prompting. Extensive experiments on three challenging benchmarks for image classification demonstrate the superiority of our model over other state-of-the-art methods for visual prompt tuning. Code is available at https://github.com/tommy-xq/SA2VP.

3D single object tracking remains a challenging problem due to the sparsity and incompleteness of the point clouds. Existing algorithms attempt to address the challenges in two strategies. The first strategy is to learn dense geometric features based on the captured sparse point cloud. Nevertheless, it is quite a formidable task since the learned dense geometric features are with high uncertainty for depicting the shape of the target object. The other strategy is to aggregate the sparse geometric features of multiple templates to enrich the shape information, which is a routine solution in 2D tracking. However, aggregating the coarse shape representations can hardly yield a precise shape representation. Different from 2D pixels, 3D points of different frames can be directly fused by coordinate transform, i.e., shape completion. Considering that, we propose to construct a synthetic target representation composed of dense and complete point clouds depicting the target shape precisely by shape completion for robust 3D tracking. Specifically, we design a voxelized 3D tracking framework with shape completion, in which we propose a quality-aware shape completion mechanism to alleviate the adverse effect of noisy historical predictions. It enables us to effectively construct and leverage the synthetic target representation. Besides, we also develop a voxelized relation modeling module and box refinement module to improve tracking performance. Favorable performance against state-of-the-art algorithms on three benchmarks demonstrates the effectiveness and generalization ability of our method.

Few-shot semantic segmentation (FSS) has achieved great success on segmenting objects of novel classes, supported by only a few annotated samples. However, existing FSS methods often underperform in the presence of domain shifts, especially when encountering new domain styles that are unseen during training. It is suboptimal to directly adapt or generalize the entire model to new domains in the few-shot scenario. Instead, our key idea is to adapt a small adapter for rectifying diverse target domain styles to the source domain. Consequently, the rectified target domain features can fittingly benefit from the well-optimized source domain segmentation model, which is intently trained on sufficient source domain data. Training domain-rectifying adapter requires sufficiently diverse target domains. We thus propose a novel local-global style perturbation method to simulate diverse potential target domains by perturbating the feature channel statistics of the individual images and collective statistics of the entire source domain, respectively. Additionally, we propose a cyclic domain alignment module to facilitate the adapter effectively rectifying domains using a reverse domain rectification supervision. The adapter is trained to rectify the image features from diverse synthesized target domains to align with the source domain. During testing on target domains, we start by rectifying the image features and then conduct few-shot segmentation on the domain-rectified features. Extensive experiments demonstrate the effectiveness of our method, achieving promising results on cross-domain few-shot semantic segmentation tasks. Our code is available at https://github.com/Matt-Su/DR-Adapter.

While Transformer has achieved remarkable performance in various high-level vision tasks, it is still challenging to exploit the full potential of Transformer in image restoration. The crux lies in the limited depth of applying Transformer in the typical encoder-decoder framework for image restoration, resulting from heavy self-attention computation load and inefficient communications across different depth (scales) of layers. In this paper, we present a deep and effective Transformer-based network for image restoration, termed as U2-Former, which is able to employ self-attention of Transformer as the core operation for feature learning to perform image restoration in a deep encoding and decoding space. Specifically, it leverages the nested U-shaped structure to facilitate the interactions across different layers with different scales of feature maps. Furthermore, we optimize the computational efficiency for the basic Transformer block by introducing a simple yet effective feature-filtering mechanism to compress the token representation. Apart from the typical supervision ways for image restoration, our U2-Former also performs multi-view contrastive learning, which constructs positive pairs in various aspects, to learn noise-sensitive but content-irrelevant features and further decouple the noise component from the background image. Extensive experiments on various image restoration tasks, including reflection removal, rain streak removal and dehazing respectively, demonstrate the effectiveness of the proposed U2-Former.

Typical Siamese-based trackers focus on the target region and pay less attention to the background area. However, the background area can provide the tracker with prior knowledge about the target surroundings. Nonetheless, since the tracker can naturally utilize the target template for localization, importing additional background knowledge requires proper design so that the background area prior knowledge can be fully explored. Furthermore, the introduction of the entire background regions is redundant. Instead, the part background distractors in the regions are more meaningful for the discrimination of the tracker. In this work, we propose a background prior knowledge fully explored tracker for robust tracking. Firstly, we present a Transformer-based explicitly and fully background-utilizing scheme by boosting the tracker to independently exploit the background for localization. Specifically, a target-distractor independent decoder explicitly utilizes the background knowledge by making the target and the distractors independently perform fusion with the search feature. Secondly, we design a simple yet efficient discriminative distractors mining module to refine the background prior knowledge by replacing the whole background region with the mined background distractors. Extensive experiments demonstrate that the proposed method performs favorably against state-of-the-art trackers on nine benchmarks.

This paper proposes a unified multi-modal image generation method to address two critical challenges in visible-infrared (VI) person re-identification (ReID): the insufficiency of training samples and the large cross-modality discrepancy. To be specific, we propose to generate cross-modal and middle-modal images to explicitly reduce the modality discrepancy, and generate intra-modal images to serve as training samples for datasets augmentation. To this end, we adapt the conditional diffusion model for multi-modal image generation. The condition includes a binary modality indicator and modal-irrelative pedestrian contour to control the target modality and pedestrian identity, respectively. For the intra-modality and cross-modality image generation, we modify the structure of UNet to take as input the conditions, and estimate the conditional probability density by optimizing its variational lower bound. Furthermore, we devise modal discriminators and adversarial training strategies to achieve modality alignment. The middle-modality image generation method shares the same network architecture with intra- and cross-modality generation, but has specific training objectives. We define the middle modality as the distribution equidistant from the visible modality and infrared modality. We employ the adversarial training to measure the distance from the visible or infrared modality to the middle modality, and thus minimize the difference between these two adversarial losses, serving as an equidistant constraint. Experimental results on SYSU-MM01 and RegDB demonstrate the effectiveness and generalization of the intra-modality, cross-modality, and middle-modality image generation.

While deep-learning-based tracking methods have achieved substantial progress, they entail large-scale and high-quality annotated data for sufficient training. To eliminate expensive and exhaustive annotation, we study self-supervised (SS) learning for visual tracking. In this work, we develop the crop-transform-paste operation, which is able to synthesize sufficient training data by simulating various appearance variations during tracking, including appearance variations of objects and background interference. Since the target state is known in all synthesized data, existing deep trackers can be trained in routine ways using the synthesized data without human annotation. The proposed target-aware data-synthesis method adapts existing tracking approaches within a SS learning framework without algorithmic changes. Thus, the proposed SS learning mechanism can be seamlessly integrated into existing tracking frameworks to perform training. Extensive experiments show that our method: 1) achieves favorable performance against supervised (Su) learning schemes under the cases with limited annotations; 2) helps deal with various tracking challenges such as object deformation, occlusion (OCC), or background clutter (BC) due to its manipulability; 3) performs favorably against the state-of-the-art unsupervised tracking methods; and 4) boosts the performance of various state-of-the-art Su learning frameworks, including SiamRPN++, DiMP, and TransT.

Real-world data usually present long-tailed distributions. Training on imbalanced data tends to render neural networks perform well on head classes while much worse on tail classes. The severe sparseness of training instances for the tail classes is the main challenge, which results in biased distribution estimation during training. Plenty of efforts have been devoted to ameliorating the challenge, including data resampling and synthesizing new training instances for tail classes. However, no prior research has exploited the transferable knowledge from head classes to tail classes for calibrating the distribution of tail classes. In this article, we suppose that tail classes can be enriched by similar head classes and propose a novel distribution calibration (DC) approach named as label-aware DC (LADC). LADC transfers the statistics from relevant head classes to infer the distribution of tail classes. Sampling from calibrated distribution further facilitates rebalancing the classifier. Experiments on both image and text long-tailed datasets demonstrate that LADC significantly outperforms existing methods. The visualization also shows that LADC provides a more accurate distribution estimation.

People may perform diverse gestures affected by various mental and physical factors when speaking the same sentences. This inherent one-to-many relationship makes co-speech gesture generation from audio particularly challenging. Conventional CNNs/RNNs assume one-to-one mapping, and thus tend to predict the average of all possible target motions, easily resulting in plain/boring motions during inference. So we propose to explicitly model the one-to-many audio-to-motion mapping by splitting the cross-modal latent code into shared code and motion-specific code. The shared code is expected to be responsible for the motion component that is more correlated to the audio while the motion-specific code is expected to capture diverse motion information that is more independent of the audio. However, splitting the latent code into two parts poses extra training difficulties. Several crucial training losses/strategies, including relaxed motion loss, bicycle constraint, and diversity loss, are designed to better train the VAE. Experiments on both 3D and 2D motion datasets verify that our method generates more realistic and diverse motions than previous state-of-the-art methods, quantitatively and qualitatively. Besides, our formulation is compatible with discrete cosine transformation (DCT) modeling and other popular backbones (i.e., RNN, Transformer). As for motion losses and quantitative motion evaluation, we find structured losses/metrics (e.g. STFT) that consider temporal and/or spatial context complement the most commonly used point-wise losses (e.g. PCK), resulting in better motion dynamics and more nuanced motion details. Finally, we demonstrate that our method can be readily used to generate motion sequences with user-specified motion clips on the timeline.

Effective image restoration with large-size corruptions, such as blind image inpainting, entails precise detection of corruption region masks which remains extremely challenging due to diverse shapes and patterns of corruptions. In this work, we present a novel method for automatic corruption detection, which allows for blind corruption restoration without known corruption masks. Specifically, we develop a hierarchical contrastive learning framework to detect corrupted regions by capturing the intrinsic semantic distinctions between corrupted and uncorrupted regions. In particular, our model detects the corrupted mask in a coarse-to-fine manner by first predicting a coarse mask by contrastive learning in low-resolution feature space and then refines the uncertain area of the mask by high-resolution contrastive learning. A specialized hierarchical interaction mechanism is designed to facilitate the knowledge propagation of contrastive learning in different scales, boosting the modeling performance substantially. The detected multi-scale corruption masks are then leveraged to guide the corruption restoration. Detecting corrupted regions by learning the contrastive distinctions rather than the semantic patterns of corruptions, our model has well generalization ability across different corruption patterns. Extensive experiments demonstrate following merits of our model: 1) the superior performance over other methods on both corruption detection and various image restoration tasks including blind inpainting and watermark removal, and 2) strong generalization across different corruption patterns such as graffiti, random noise or other image content. Codes and trained weights are available at https://github.com/xyfJASON/HCL.

Although extensive research has been conducted on 3D point cloud segmentation, effectively adapting generic models to novel categories remains a formidable challenge. This paper proposes a novel approach to improve point cloud few-shot segmentation (PC-FSS) models. Unlike existing PC-FSS methods that directly utilize categorical information from support prototypes to recognize novel classes in query samples, our method identifies two critical aspects that substantially enhance model performance by reducing contextual gaps between support prototypes and query features. Specifically, we (1) adapt support background prototypes to match query context while removing extraneous cues that may obscure foreground and background in query samples, and (2) holistically rectify support prototypes under the guidance of query features to emulate the latter having no semantic gap to the query targets. Our proposed designs are agnostic to the feature extractor, rendering them readily applicable to any prototype-based methods. The experimental results on S3DIS and ScanNet demonstrate notable practical benefits, as our approach achieves significant improvements while still maintaining high efficiency. The code for our approach is available at https://github.com/AaronNZH/Boosting-Few-shot-3D-Point-Cloud-Segmentation-via-Query-Guided-Enhancement

Existing methods for interactive segmentation in radiance fields entail scene-specific optimization and thus cannot generalize across different scenes, which greatly limits their applicability. In this work we make the first attempt at Scene-Generalizable Interactive Segmentation in Radiance Fields (SGISRF) and propose a novel SGISRF method, which can perform 3D object segmentation for novel (unseen) scenes represented by radiance fields, guided by only a few interactive user clicks in a given set of multi-view 2D images. In particular, the proposed SGISRF focuses on addressing three crucial challenges with three specially designed techniques. First, we devise the Cross-Dimension Guidance Propagation to encode the scarce 2D user clicks into informative 3D guidance representations. Second, the Uncertainty-Eliminated 3D Segmentation module is designed to achieve efficient yet effective 3D segmentation. Third, Concealment-Revealed Supervised Learning scheme is proposed to reveal and correct the concealed 3D segmentation errors resulted from the supervision in 2D space with only 2D mask annotations. Extensive experiments on two real-world challenging benchmarks covering diverse scenes demonstrate 1) effectiveness and scene-generalizability of the proposed method, 2) favorable performance compared to classical method requiring scene-specific optimization.

Recent interactive segmentation methods iteratively take source image, user guidance and previously predicted mask as the input without considering the invariant nature of the source image. As a result, the process of extracting features from the source image is repeated in each interaction, resulting in substantial computational redundancy. In this work, we propose the Feature Decoupling-Recycling Network (FDRN), which decouples the modeling components based on their intrinsic discrepancies and then recycles components that can be reused for each user interaction. Thus, the efficiency of the whole interactive process can be significantly improved. To be specific, we apply the Decoupling-Recycling strategy from three perspectives to address three types of discrepancies, respectively. First, our model decouples the learning of source image semantics from the encoding of user guidance to process two types of input domains separately. Second, FDRN decouples high-level and low-level features from stratified semantic representations to enhance feature learning. Third, during the encoding of user guidance, current user guidance is decoupled from historical guidance to highlight the effect of current user guidance. We conduct extensive experiments on 6 datasets from different domains and modalities, which demonstrate the following merits of our model: 1) superior efficiency than other methods, particularly advantageous in the challenging scenarios requiring long-term interactions (up to 4.25x faster), while achieving favorable segmentation performance; 2) strong applicability to various methods serving as a universal enhancement technique; 3) well cross-task generalizability, e.g., to medical image segmentation, and robustness against misleading user guidance.

The crux of text-to-image synthesis stems from the difficulty of preserving the cross-modality semantic consistency between the input text and the synthesized image. Typical methods, which seek to model the text-to-image mapping directly, could only capture keywords in the text that indicates common objects or actions but fail to learn their spatial distribution patterns. An effective way to circumvent this limitation is to generate an image layout as guidance, which is attempted by a few methods. Nevertheless, these methods fail to generate practically effective layouts due to the diversity of input text and object location. In this paper we push for effective modeling in both text-to-layout generation and layout-to-image synthesis. Specifically, we formulate the text-to-layout generation as a sequence-to-sequence modeling task, and build our model upon Transformer to learn the spatial relationships between objects by modeling the sequential dependencies between them. In the stage of layout-to-image synthesis, we focus on learning the textual-visual semantic alignment per object in the layout to precisely incorporate the input text into the layout-to-image synthesizing process. To evaluate the quality of generated layout, we design a new metric specifically, dubbed Layout Quality Score, which considers both the absolute distribution errors of bounding boxes in the layout and the mutual spatial relationships between them. Extensive experiments on three datasets demonstrate the superior performance of our method over state-of-the-art methods on both predicting the layout and synthesizing the image from the given text.

Cross-modal image-text retrieval is an important area of Vision-and-Language task that models the similarity of image-text pairs by embedding features into a shared space for alignment. To bridge the heterogeneous gap between the two modalities, current approaches achieve inter-modal alignment and intra-modal semantic relationship modeling through complex weighted combinations between items. In the intra-modal association and inter-modal interaction processes, the higher-weight items have a higher contribution to the global semantics. However, the same item always produces different contributions in the two processes, since most traditional approaches only focus on the alignment. This usually results in semantic changes and misalignment. To address this issue, this paper proposes Cross-modal Semantic Importance Consistency (CSIC) which achieves invariance in the semantic of items during aligning. The proposed technique measures the semantic importance of items obtained from intra-modal and inter-modal self-attention and learns a more reasonable representation vector by inter-calibrating the importance distribution to improve performance. We conducted extensive experiments on the Flickr30K and MS COCO datasets. The results show that our approach can significantly improve retrieval performance, proving the proposed approach’s superiority and rationality.

Typical methods for pedestrian detection focus on either tackling mutual occlusions between crowded pedestrians, or dealing with the various scales of pedestrians. Detecting pedestrians with substantial appearance diversities such as different pedestrian silhouettes, different viewpoints or different dressing, remains a crucial challenge. Instead of learning each of these diverse pedestrian appearance features individually as most existing methods do, we propose to perform contrastive learning to guide the feature learning in such a way that the semantic distance between pedestrians with different appearances in the learned feature space is minimized to eliminate the appearance diversities, whilst the distance between pedestrians and background is maximized. To facilitate the efficiency and effectiveness of contrastive learning, we construct an exemplar dictionary with representative pedestrian appearances as prior knowledge to construct effective contrastive training pairs and thus guide contrastive learning. Besides, the constructed exemplar dictionary is further leveraged to evaluate the quality of pedestrian proposals during inference by measuring the semantic distance between the proposal and the exemplar dictionary. Extensive experiments on both daytime and nighttime pedestrian detection validate the effectiveness of the proposed method.

Bin Liang, Chenwei Lou, Xiang Li, Min Yang, Lin Gui, Yulan He, Wenjie Pei, Ruifeng Xu. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

The crux of long-term tracking lies in the difficulty of tracking the target with discontinuous moving caused by out-of-view or occlusion. Existing long-term tracking methods follow two typical strategies. The first strategy employs a local tracker to perform smooth tracking and uses another re-detector to detect the target when the target is lost. While it can exploit the temporal context like historical appearances and locations of the target, a potential limitation of such strategy is that the local tracker tends to misidentify a nearby distractor as the target instead of activating the re-detector when the real target is out of view. The other long-term tracking strategy tracks the target in the entire image globally instead of local tracking based on the previous tracking results. Unfortunately, such global tracking strategy cannot leverage the temporal context effectively. In this work, we combine the advantages of both strategies: tracking the target in a global view while exploiting the temporal context. Specifically, we perform global tracking via ensemble of local trackers spreading the full image. The smooth moving of the target can be handled steadily by one local tracker. When the local tracker accidentally loses the target due to suddenly discontinuous moving, another local tracker close to the target is then activated and can readily take over the tracking to locate the target. While the activated local tracker performs tracking locally by leveraging the temporal context, the ensemble of local trackers renders our model the global view for tracking. Extensive experiments on six datasets demonstrate that our method performs favorably against state-of-the-art algorithms.

Typical methods for blind image super-resolution (SR) focus on dealing with unknown degradations by directly estimating them or learning the degradation representations in a latent space. A potential limitation of these methods is that they assume the unknown degradations can be simulated by the integration of various handcrafted degradations (e.g., bicubic downsampling), which is not necessarily true. The real-world degradations can be beyond the simulation scope by the handcrafted degradations, which are referred to as novel degradations. In this work, we propose to learn a latent representation space for degradations, which can be generalized from handcrafted (base) degradations to novel degradations. Furthermore, we perform variational inference to match the posterior of degradations in latent representation space with a prior distribution (e.g., Gaussian distribution). Consequently, we are able to sample more high-quality representations for a novel degradation to augment the training data for SR model. We conduct extensive experiments on both synthetic and real-world datasets to validate the effectiveness and advantages of our method for blind super-resolution with novel degradations.

Typical text spotters follow the two-stage spotting strategy: detect the precise boundary for a text instance first and then perform text recognition within the located text region. While such strategy has achieved substantial progress, there are two underlying limitations. 1) The performance of text recognition depends heavily on the precision of text detection, resulting in the potential error propagation from detection to recognition. 2) The RoI cropping which bridges the detection and recognition brings noise from background and leads to information loss when pooling or interpolating from feature maps. In this work we propose the single shot Self-Reliant Scene Text Spotter (SRSTS), which circumvents these limitations by decoupling recognition from detection. Specifically, we conduct text detection and recognition in parallel and bridge them by the shared positive anchor point. Consequently, our method is able to recognize the text instances correctly even though the precise text boundaries are challenging to detect. Additionally, our method reduces the annotation cost for text detection substantially. Extensive experiments on regular-shaped benchmark and arbitrary-shaped benchmark demonstrate that our SRSTS compares favorably to previous state-of-the-art spotters in terms of both accuracy and efficiency.

Few-shot learning (FSL) targets at generalization of vision models towards unseen tasks without sufficient annotations. Despite the emergence of a number of few-shot learning methods, the sample selection bias problem, i.e., the sensitivity to the limited amount of support data, has not been well understood. In this paper, we find that this problem usually occurs when the positions of support samples are in the vicinity of task centroid—the mean of all class centroids in the task. This motivates us to propose an extremely simple feature transformation to alleviate this problem, dubbed Task Centroid Projection Removing (TCPR). TCPR is applied directly to all image features in a given task, aiming at removing the dimension of features along the direction of the task centroid. While the exact task centoid cannot be accurately obtained from limited data, we estimate it using base features that are each similar to one of the support features. Our method effectively prevents features from being too close to the task centroid. Extensive experiments over ten datasets from different domains show that TCPR can reliably improve classification accuracy across various feature extractors, training algorithms and datasets. The code has been made available at https://github.com/KikimorMay/FSL-TCBR.

The crux of single-channel speech separation is how to encode the mixture of signals into such a latent embedding space that the signals from different speakers can be precisely separated. Existing methods for speech separation either transform the speech signals into frequency domain to perform separation or seek to learn a separable embedding space by constructing a latent domain based on convolutional filters. While the latter type of methods learning an embedding space achieves substantial improvement for speech separation, we argue that the embedding space defined by only one latent domain does not suffice to provide a thoroughly separable encoding space for speech separation. In this paper, we propose the Stepwise-Refining Speech Separation Network (SRSSN), which follows a coarse-to-fine separation framework. It first learns a 1-order latent domain to define an encoding space and thereby performs a rough separation in the coarse phase. Then the proposedSRSSNlearns a new latent domain along each basis function of the existing latent domain to obtain a high-order latent domain in the refining phase, which enables our model to perform a refining separation to achieve a more precise speech separation. We demonstrate the effectiveness of ourSRSSNby conducting extensive experiments, including speech separation in a clean (noise-free) setting on WSJ0-2/3mix datasets as well as in noisy/reverberant settings on WHAM!/WHAMR! datasets. Furthermore, we also perform experiments of speech recognition on separated speech signals by our model to evaluate the performance of speech separation indirectly.

The critical challenge of single image inpainting stems from accurate semantic inference via limited information while maintaining image quality. Typical methods for semantic image inpainting train an encoder-decoder network by learning a one-to-one mapping from the corrupted image to the inpainted version. While such methods perform well on images with small corrupted regions, it is challenging for these methods to deal with images with large corrupted area due to two potential limitations. 1) Such one-to-one mapping paradigm tends to overfit each single training pair of images; 2) The inter-image prior knowledge about the general distribution patterns of visual semantics, which can be transferred across images sharing similar semantics, is not explicitly exploited. In this paper, we propose the Generative Memory-guided Semantic Reasoning Model (GM-SRM), which infers the content of corrupted regions based on not only the known regions of the corrupted image, but also the learned inter-image reasoning priors characterizing the generalizable semantic distribution patterns between similar images. In particular, the proposed GM-SRM first pre-learns a generative memory from the whole training data to explicitly learn the distribution of different semantic patterns. Then the learned memory are leveraged to retrieve the matching semantics for the current corrupted image to perform semantic reasoning during image inpainting. While the encoder-decoder network is used for guaranteeing the pixel-level content consistency, our generative priors are favorable for performing high-level semantic reasoning, which is particularly effective for inferring semantic content for large corrupted area. Extensive experiments on Paris Street View, CelebA-HQ, and Places2 benchmarks demonstrate that our GM-SRM outperforms the state-of-the-art methods for image inpainting in terms of both visual quality and quantitative metrics.

The current Siamese network based on region proposal network (RPN) has attracted great attention in visual tracking due to its excellent accuracy and high efficiency. However, the design of the RPN involves the selection of the number, scale, and aspect ratios of anchor boxes, which will affect the applicability and convenience of the model. Furthermore, these anchor boxes require complicated calculations, such as calculating their intersection-over-union (IoU) with ground truth bounding boxes. Due to the problems related to anchor boxes, we propose a simple yet effective anchor-free tracker (named Siamese corner networks, SiamCorners), which is end-to-end trained offline on large-scale image pairs. Specifically, we introduce a modified corner pooling layer to convert the bounding box estimate of the target into a pair of corner predictions (the bottom-right and the top-left corners). By tracking a target as a pair of corners, we avoid the need to design the anchor boxes. This will make the entire tracking algorithm more flexible and simple than anchor-based trackers. In our network design, we further introduce a layer-wise feature aggregation strategy that enables the corner pooling module to predict multiple corners for a tracking target in deep networks. We then introduce a new penalty term that is used to select an optimal tracking box in these candidate corners. Finally, SiamCorners achieves experimental results that are comparable to the state-of-art tracker while maintaining a high running speed. In particular, SiamCorners achieves a 53.7% AUC on NFS30 and a 61.4% AUC on UAV123, while still running at 42 frames per second (FPS).

Generating conversational gestures from speech audio is challenging due to the inherent one-to-many mapping be-tween audio and body motions. Conventional CNNs/RNNs assume one-to-one mapping, and thus tend to predict the average of all possible target motions, resulting in plain/boring motions during inference. In order to over-come this problem, we propose a novel conditional variational autoencoder (VAE) that explicitly models one-to-many audio-to-motion mapping by splitting the cross-modal latent code into shared code and motion-specific code. The shared code mainly models the strong correlation between audio and motion (such as the synchronized audio and motion beats), while the motion-specific code captures diverse motion information independent of the audio. However, splitting the latent code into two parts poses training difficulties for the VAE model. A mapping network facilitating random sampling along with other techniques including relaxed motion loss, bicycle constraint, and diversity loss are designed to better train the VAE. Experiments on both 3D and 2D motion datasets verify that our method generates more realistic and diverse motions than state-of-the-art methods, quantitatively and qualitatively. Finally, we demonstrate that our method can be readily used to generate motion sequences with user-specified motion clips on the timeline. Code and more results are at https://jingli513.github.io/audio2gestures.

Most existing trackers based on deep learning perform tracking in a holistic strategy, which aims to learn deep representations of the whole target for localizing the target. It is arduous for such methods to track targets with various appearance variations. To address this limitation, another type of methods adopts a part-based tracking strategy which divides the target into equal patches and tracks all these patches in parallel. The target state is inferred by summarizing the tracking results of these patches. A potential limitation of such trackers is that not all patches are equally informative for tracking. Some patches that are not discriminative may have adverse effects. In this paper, we propose to track the salient local parts of the target that are discriminative for tracking. In particular, we propose a fine-grained saliency mining module to capture the local saliencies. Further, we design a saliency-association modeling module to associate the captured saliencies together to learn effective correlation representations between the exemplar and the search image for state estimation. Extensive experiments on five diverse datasets demonstrate that the proposed method performs favorably against state-of-the-art trackers.

The crux of image reflection removal stems from the difficulty of recognizing the diverse reflection patterns. Typical methods optimize the modeling of background restoration by performing low-level supervision on the restored image to minimize its per-pixel difference from the groundtruth, which re-lies on substantial training samples to learn diverse reflection patterns robustly and avoid overfitting spurious reflection patterns. In this work, we perform supervision on the contrastive distribution between the predicted background and the reflection image. Specifically, our proposed method restores the background and the reflection images in parallel, and seeks to maximize the distribution consistency between the predicted background-reflection contrast and the groundtruth contrast in the latent space. Such supervision pushes the model to focus on contrastive modeling between the background and reflection image. Extensive experiments on four real-world bench-marks demonstrate that our method consistently outperforms state-of-the-art methods.

Restoring the clean background from the superimposed images containing a noisy layer is the common crux of a classical category of tasks on image restoration such as image reflection removal, image deraining and image dehazing. These tasks are typically formulated and tackled individually due to diverse and complicated appearance patterns of noise layers within the image. In this work we present the Deep-Masking Generative Network (DMGN), which is a unified framework for background restoration from the superimposed images and is able to cope with different types of noise. Our proposed DMGN follows a coarse-to-fine generative process: a coarse background image and a noise image are first generated in parallel, then the noise image is further leveraged to refine the background image to achieve a higher-quality background image. In particular, we design the novel Residual Deep-Masking Cell as the core operating unit for our DMGN to enhance the effective information and suppress the negative information during image generation via learning a gating mask to control the information flow. By iteratively employing this Residual Deep-Masking Cell, our proposed DMGN is able to generate both high-quality background image and noisy image progressively. Furthermore, we propose a two-pronged strategy to effectively leverage the generated noise image as contrasting cues to facilitate the refinement of the background image. Extensive experiments across three typical tasks for image background restoration, including image reflection removal, image rain steak removal and image dehazing, show that our DMGN consistently outperforms state-of-the-art methods specifically designed for each single task.

Person re-identification aims to identify whether pairs of images belong to the same person or not. This problem is challenging due to large differences in camera views, lighting and background. One of the mainstream in learning CNN features is to design loss functions which reinforce both the class separation and intra-class compactness. In this paper, we propose a novel Orthogonal Center Learning method with Subspace Masking for person re-identification. We make the following contributions: 1) we develop a center learning module to learn the class centers by simultaneously reducing the intra-class differences and inter-class correlations by orthogonalization; 2) we introduce a subspace masking mechanism to enhance the generalization of the learned class centers; and 3) we propose to integrate the average pooling and max pooling in a regularizing manner that fully exploits their powers. Extensive experiments show that our proposed method consistently outperforms the state-of-the-art methods on large-scale ReID datasets including Market-1501, DukeMTMC-ReID, CUHK03 and MSMT17.

The crux of homography estimation is that the homography is characterized by the geometric correspondences between two related images rather than appearance features, which differs from typical image recognition tasks. Existing methods either decompose the task of homography estimation into several individual sub-problems and optimize them sequentially, or attempt to tackle it in an end-to-end manner by delegating the whole task to deep convolutional networks (CNNs). However, it is quite arduous for CNNs to learn the mapping function from appearance features of related images to the homography directly. In this paper, we propose to parse the geometric correspondences between related images explicitly to bridge the gap between deep appearance features and the homography. Furthermore, we propose a coarse-to-fine estimation framework to capture different scale of homography transformations and thus predict the homography in a stepwise-refining manner. Additionally, we propose a pyramidal supervision scheme to leverage an important prior concerning the homography estimation. Extensive experiments on two large-scale datasets demonstrate that our model advances the state-of-the-art performance significantly.

The main challenges of age estimation from facial expression videos lie not only in the modeling of the static facial appearance, but also in the capturing of the temporal facial dynamics. Traditional techniques to this problem focus on constructing handcrafted features to explore the discriminative information contained in facial appearance and dynamics separately. This relies on sophisticated feature-refinement and framework-design. In this paper, we present an end-toend architecture for age estimation, called Spatially-Indexed Attention Model (SIAM), which is able to simultaneously learn both the appearance and dynamics of age from raw videos of facial expressions. Specifically, we employ convolutional neural networks to extract effective latent appearance representations and feed them into recurrent networks to model the temporal dynamics. More importantly, we propose to leverage attention models for salience detection in both the spatial domain for each single image and the temporal domain for the whole video as well. We design a specific spatially-indexed attention mechanism among the convolutional layers to extract the salient facial regions in each individual image, and a temporal attention layer to assign attention weights to each frame. This two-pronged approach not only improves the performance by allowing the model to focus on informative frames and facial areas, but it also offers an interpretable correspondence between the spatial facial regions as well as temporal frames, and the task of age estimation. We demonstrate the strong performance of our model in experiments on a large, gender-balanced database with 400 subjects with ages spanning from 8 to 76 years. Experiments reveal that our model exhibits significant superiority over the state-of-the-art methods given sufficient training data.

Typical techniques for video captioning follow the encoder-decoder framework, which can only focus on one source video being processed. A potential disadvantage of such design is that it cannot capture the multiple visual context information of a word appearing in more than one relevant videos in training data. To tackle this limitation, we propose the Memory-Attended Recurrent Network (MARN) for video captioning, in which a memory structure is designed to explore the full-spectrum correspondence between a word and its various similar visual contexts across videos in training data. Thus, our model is able to achieve a more comprehensive understanding for each word and yield higher captioning quality. Furthermore, the built memory structure enables our method to model the compatibility between adjacent words explicitly instead of asking the model to learn implicitly, as most existing models do. Extensive validation on two real-word datasets demonstrates that our MARN consistently outperforms state-of-the-art methods.

Typical methods for supervised sequence modeling are built upon the recurrent neural networks to capture temporal dependencies. One potential limitation of these methods is that they only model explicitly information interactions between adjacent time steps in a sequence, hence the high-order interactions between nonadjacent time steps are not fully exploited. It greatly limits the capability of modeling the long-range temporal dependencies since one-order interactions cannot be maintained for a long term due to information dilution and gradient vanishing. To tackle this limitation, we propose the Non-local Recurrent Neural Memory (NRNM) for supervised sequence modeling, which performs non-local operations to learn full-order interactions within a sliding temporal block and models the global interactions between blocks in a gated recurrent manner. Consequently, our model is able to capture the long-range dependencies. Besides, the latent high-level features contained in high-order interactions can be distilled by our model. We demonstrate the merits of our NRNM approach on two different tasks: action recognition and sentiment analysis.

State-of-the-art image captioning methods mostly focus on improving visual features, less attention has been paid to utilizing the inherent properties of language to boost captioning performance. In this paper, we show that vocabulary coherence between words and syntactic paradigm of sentences are also important to generate high-quality image caption. Following the conventional encoder-decoder framework, we propose the Reflective Decoding Network (RDN) for image captioning, which enhances both the long-sequence dependency and position perception of words in a caption decoder. Our model learns to collaboratively attend on both visual and textual features and meanwhile perceive each word's relative position in the sentence to maximize the information delivered in the generated caption. We evaluate the effectiveness of our RDN on the COCO image captioning datasets and achieve superior performance over the previous methods. Further experiments reveal that our approach is particularly advantageous for hard cases with complex scenes to describe by captions.

We present a new model for multivariate time-series classification, called the hidden-unit logistic model (HULM), that uses binary stochastic hidden units to model latent structure in the data. The hidden units are connected in a chain structure that models temporal dependencies in the data. Compared with the prior models for time-series classification such as the hidden conditional random field, our model can model very complex decision boundaries, because the number of latent states grows exponentially with the number of hidden units. We demonstrate the strong performance of our model in experiments on a variety of (computer vision) tasks, including handwritten character recognition, speech recognition, facial expression, and action recognition. We also present a state-of-the-art system for facial action unit detection based on the HULM.

Capturing the temporal dynamics of user preferences over items is important for recommendation. Existing methods mainly assume that all time steps in user-item interaction history are equally relevant to recommendation, which however does not apply in real-world scenarios where user-item interactions can often happen accidentally. More importantly, they learn user and item dynamics separately, thus failing to capture their joint effects on user-item interactions. To better model user and item dynamics, we present the Interacting Attention-gated Recurrent Network (IARN) which adopts the attention model to measure the relevance of each time step. In particular, we propose a novel attention scheme to learn the attention scores of user and item history in an interacting way, thus to account for the dependencies between user and item dynamics in shaping user-item interactions. By doing so, IARN can selectively memorize different time steps of a user's history when predicting her preferences over different items. Our model can therefore provide meaningful interpretations for recommendation results, which could be further enhanced by auxiliary features. Extensive validation on real-world datasets shows that IARN consistently outperforms state-of-the-art methods.

Typical techniques for sequence classification are designed for well-segmented sequences which have been edited to remove noisy or irrelevant parts. Therefore, such methods cannot be easily applied on noisy sequences expected in real-world applications. In this paper, we present the Temporal Attention-Gated Model (TAGM) which integrates ideas from attention models and gated recurrent networks to better deal with noisy or unsegmented sequences. Specifically, we extend the concept of attention model to measure the relevance of each observation (time step) of a sequence. We then use a novel gated recurrent network to learn the hidden representation for the final prediction. An important advantage of our approach is interpretability since the temporal attention weights provide a meaningful value for the salience of each time step in the sequence. We demonstrate the merits of our TAGM approach, both for prediction accuracy and interpretability, on three different tasks: spoken digit recognition, text-based sentiment analysis and visual event recognition.

Image-space line integral convolution (LIC) is a popular approach for visualizing surface vector fields due to its simplicity and high efficiency. To avoid inconsistencies or color blur during the user interactions in the image-space approach, some methods use surface parameterization or 3D volume texture for the effect of smooth transition, which often require expensive computational or memory cost. Furthermore, those methods cannot achieve consistent LIC results in both granularity and color distribution on different scales. This paper introduces a novel image-space LIC for surface flows that preserves the texture coherence during user interactions. To make the noise textures under different viewpoints coherent, we propose a simple texture mapping technique that is local, robust and effective. Meanwhile, our approach pre-computes a sequence of mipmap noise textures in a coarse-to-fine manner, leading to consistent transition when the model is zoomed. Prior to perform LIC in the image space, the mipmap noise textures are mapped onto each triangle with randomly assigned texture coordinates. Then, a standard image-space LIC based on the projected vector fields is performed to generate the flow texture. The proposed approach is simple and very suitable for GPU acceleration. Our implementation demonstrates consistent and highly efficient LIC visualization on a variety of datasets.