Traditional drug discovery is a time-consuming and costly endeavor. To address this challenge, this research developed a deep-learning-based conditional generative model. This model is designed to generate small molecules guided by specific protein sequences (e.g., EGFR), endowing them with targeted pharmacological activity, high novelty, uniqueness, and promising drug-likeness, thereby accelerating pan-cancer drug molecule design. The study employed a hybrid architecture combining Graph Neural Networks (GNNs) for processing molecular structures and a Transformer model for encoding protein sequences, which together provide conditional guidance for the generation process. The model's performance was evaluated on a validation set using metrics such as training loss curves, Mean Squared Error (MSE), and$\mathrm{R}^{2}$(coefficient of determination). The quality of the generated molecules was assessed based on their validity, uniqueness, novelty, and the number of violations against Lipinski's Rule of Five. The results indicate that the model training was stable and convergent, demonstrating good generalization ability. The generated molecules exhibited excellent novelty (100%) and drug-likeness (with an average of zero Lipinski's rule violations), while achieving 100% uniqueness. Although the validity rate was 22%, suggesting room for improvement, these findings collectively underscore the model's significant potential for conditional innovative drug molecule design. This work lays a solid foundation for future model optimization to enhance generation efficiency and explore its broader applications in drug discovery.

In this study, we introduce the concept of creative agents, aiming to enhance the capabilities of large language models (LLMs) in tasks related to the generation of prototypes of creative works. Our approach utilizes both self-collaboration and self-correction mechanisms, facilitated by hierarchical agents, to address the inherent complexities of target tasks. Our key observations are two-fold: first, effective task solving demands in-depth domain knowledge and intricate reasoning, for which deploying specialized agents for individual sub-tasks can markedly enhance LLM performance. Second, task solving intrinsically adheres to a hierarchical execution structure, comprising both high-level strategic planning and detailed task execution. Towards this end, our Creative-Agent paradigm aligns closely with this hierarchical structure, promising enhanced efficacy and adaptability across a wide range of scenarios. Specifically, our framework includes two novel modules, i.e., hierarchical Objects and Key Results generation and multi-level evaluation, both contributing to more efficient and robust task-solving. In practice, hierarchical OKR generation decomposes objects into multiple sub-objects and assigns new agents based on key results and agent responsibilities. These agents subsequently elaborate on their designated tasks and may further decompose them as necessary. Such generation operates recursively and hierarchically, culminating in a comprehensive set of detailed solutions. The multi-level evaluation module of Creative-Agent refines the solution by leveraging feedback from all associated agents, optimizing each step of the process. This scheme ensures the solution is accurate, practical, and meets the requirements of intricate tasks, enhancing the overall reliability and quality of the outcome. Experimental results demonstrated the effectiveness of Creative-Agent in completing the task of creating prototype generation works.

Subject-to-Video (S2V) generation aims to create videos that faithfully incorporate reference content, providing enhanced flexibility in the production of videos. To establish the infrastructure for S2V generation, we propose OpenS2V-Nexus, consisting of (i) OpenS2V‑Eval, a fine‑grained benchmark, and (ii) OpenS2V‑5M, a million‑scale dataset.In contrast to existing S2V benchmarks inherited from VBench that focus on global and coarse-grained assessment of generated videos, OpenS2V-Eval focuses on the model's ability to generate subject-consistent videos with natural subject appearance and identity fidelity. For these purposes, OpenS2V-Eval introduces 180 prompts from seven major categories of S2V, which incorporate both real and synthetic test data. Furthermore, to accurately align human preferences with S2V benchmarks, we propose three automatic metrics, NexusScore, NaturalScore and GmeScore, to separately quantify subject consistency, naturalness, and text relevance in generated videos. Building on this, we conduct a comprehensive evaluation of 18 representative S2V models, highlighting their strengths and weaknesses across different content. Moreover, we create the first open-source large-scale S2V generation dataset OpenS2V-5M, which consists of five million high-quality 720P subject-text-video triplets. Specifically, we ensure subject‐information diversity in our dataset by (1) segmenting subjects and building pairing information via cross‐video associations and (2) prompting GPT-4o on raw frames to synthesize multi-view representations. Through OpenS2V-Nexus, we deliver a robust infrastructure to accelerate future S2V generation research.

Despite the impressive results of arbitrary image-guided style transfer methods, text-driven image stylization has recently been proposed for transferring a natural image into a stylized one according to textual descriptions of the target style provided by the user. Unlike the previous image-to-image transfer approaches, text-guided stylization progress provides users with a more precise and intuitive way to express the desired style. However, the huge discrepancy between cross-modal inputs/outputs makes it challenging to conduct text-driven image stylization in a typical feed-forward CNN pipeline. In this article, we present DiffStyler, a dual diffusion processing architecture to control the balance between the content and style of the diffused results. The cross-modal style information can be easily integrated as guidance during the diffusion process step-by-step. Furthermore, we propose a content image-based learnable noise on which the reverse denoising process is based, enabling the stylization results to better preserve the structure information of the content image. We validate the proposed DiffStyler beyond the baseline methods through extensive qualitative and quantitative experiments. The code is available at https://github.com/haha-lisa/Diffstyler.

Although remarkable progress has been made in image style transfer, style is just one of the components of artistic paintings. Directly transferring extracted style features to natural images often results in outputs with obvious synthetic traces. This is because key painting attributes including layout, perspective, shape, and semantics often cannot be conveyed and expressed through style transfer. Large-scale pretrained text-to-image generation models have demonstrated their capability to synthesize a vast amount of high-quality images. However, even with extensive textual descriptions, it is challenging to fully express the unique visual properties and details of paintings. Moreover, generic models often disrupt the overall artistic effect when modifying specific areas, making it more complicated to achieve a unified aesthetic in artworks. Our main novel idea is to integrate multimodal semantic information as a synthesis guide into artworks, rather than transferring style to the real world. We also aim to reduce the disruption to the harmony of artworks while simplifying the guidance conditions. Specifically, we propose an innovative multi-task unified framework called CreativeSynth, based on the diffusion model with the ability to coordinate multimodal inputs. CreativeSynth combines multimodal features with customized attention mechanisms to seamlessly integrate real-world semantic content into the art domain through Cross-Art-Attention for aesthetic maintenance and semantic fusion. We demonstrate the results of our method across a wide range of different art categories, proving that CreativeSynth bridges the gap between generative models and artistic expression.

The essence of a video lies in the dynamic motions. While text-to-video generative diffusion models have made significant strides in creating diverse content, effectively controlling specific motions through text prompts remains a challenge. By utilizing user-specified reference videos, the more precise guidance for character actions, object movements, and camera movements can be achieved. This gives rise to the task of motion customization, where the primary challenge lies in effectively decoupling the appearance and motion within a video clip. To address this challenge, we introduce MotionCrafter, a novel one-shot instance-guided motion customization method that is suitable for both pre-trained text-to-video and text-to-image diffusion models. MotionCrafter employs a parallel spatial-temporal architecture that integrates the reference motion into the temporal component of the base model, while independently adjusting the spatial module for character or style control. To enhance the disentanglement of motion and appearance, we propose an innovative dual-branch motion disentanglement approach, which includes a motion disentanglement loss and an appearance prior enhancement strategy. To facilitate more efficient learning of motions, we further propose a novel timestep-layered tuning strategy that directs the diffusion model to focus on motion-level information. Through comprehensive quantitative and qualitative experiments, along with user preference tests, we demonstrate that MotionCrafter can successfully integrate dynamic motions while maintaining the coherence and quality of the base model, providing a wide range of appearance generation capabilities. MotionCrafter can be applied to various personalized backbones in the community to generate videos with a variety of artistic styles.

With the development of diffusion models, text-guided image style transfer has demonstrated great controllable and high-quality results. However, the utilization of text for diverse music style transfer poses significant challenges, primarily due to the limited availability of matched audio-text datasets. Music, being an abstract and complex art form, exhibits variations and intricacies even within the same genre, thereby making accurate textual descriptions challenging. This paper presents a music style transfer approach that effectively captures musical attributes using minimal data. We introduce a novel time-varying textual inversion module to precisely capture mel-spectrogram features at different levels. During inference, we utilize a bias-reduced stylization technique to get stable results. Experimental results demonstrate that our method can transfer the style of specific instruments, as well as incorporate natural sounds to compose melodies. Samples and code are available at https://lsfhuihuiff.github.io/MusicTI/.

Current techniques for detecting AI-generated text are largely confined to manual feature crafting and supervised binary classification paradigms. These methodologies typically lead to performance bottlenecks and unsatisfactory generalizability. Consequently, these methods are often inapplicable for out-of-distribution (OOD) data and newly emerged large language models (LLMs). In this paper, we revisit the task of AI-generated text detection. We argue that the key to accomplishing this task lies in distinguishing writing styles of different authors, rather than simply classifying the text into human-written or AI-generated text. To this end, we propose DeTeCtive, a multi-task auxiliary, multi-level contrastive learning framework. DeTeCtive is designed to facilitate the learning of distinct writing styles, combined with a dense information retrieval pipeline for AI-generated text detection. Our method is compatible with a range of text encoders. Extensive experiments demonstrate that our method enhances the ability of various text encoders in detecting AI-generated text across multiple benchmarks and achieves state-of-the-art results. Notably, in OOD zero-shot evaluation, our method outperforms existing approaches by a large margin. Moreover, we find our method boasts a Training-Free Incremental Adaptation (TFIA) capability towards OOD data, further enhancing its efficacy in OOD detection scenarios. We will open-source our code and models in hopes that our work will spark new thoughts in the field of AI-generated text detection, ensuring safe application of LLMs and enhancing compliance.

The seamless integration of music with dance movements is essential for communicating the artistic intent of a dance piece. This alignment also significantly improves the immersive quality of gaming experiences and animation productions. Although there has been remarkable advancement in creating high-fidelity music from textual descriptions, current methodologies mainly focus on modulating overall characteristics such as genre and emotional tone. They often overlook the nuanced management of temporal rhythm, which is indispensable in crafting music for dance, since it intricately aligns the musical beats with the dancers’ movements. Recognizing this gap, we propose an encoder-based textual inversion technique to augment text-to-music models with visual control, facilitating personalized music generation. Specifically, we develop dual-path rhythm-genre inversion to effectively integrate the rhythm and genre of a dance motion sequence into the textual space of a text-to-music model. Contrary to traditional textual inversion methods, which directly update text embeddings to reconstruct a single target object, our approach utilizes separate rhythm and genre encoders to obtain text embeddings for two pseudo-words, adapting to the varying rhythms and genres. We collect a new dataset called In-the-wild Dance Videos (InDV) and demonstrate that our approach outperforms state-of-the-art methods across multiple evaluation metrics. Furthermore, our method is able to adapt to changes in tempo and effectively integrates with the inherent text-guided generation capability of the pre-trained model. Our source code and demo videos are available at https://github.com/lsfhuihuiff/Dance-to-music_Siggraph_Asia_2024.

Arbitrary image stylization by neural networks has become a popular topic, and video stylization is attracting more attention as an extension of image stylization. However, when image stylization methods are applied to videos, unsatisfactory results that suffer from severe flickering effects appear. In this article, we conducted a detailed and comprehensive analysis of the cause of such flickering effects. Systematic comparisons among typical neural style transfer approaches show that the feature migration modules for state-of-the-art (SOTA) learning systems are ill-conditioned and could lead to a channelwise misalignment between the input content representations and the generated frames. Unlike traditional methods that relieve the misalignment via additional optical flow constraints or regularization modules, we focus on keeping the temporal consistency by aligning each output frame with the input frame. To this end, we propose a simple yet efficient multichannel correlation network (MCCNet), to ensure that output frames are directly aligned with inputs in the hidden feature space while maintaining the desired style patterns. An inner channel similarity loss is adopted to eliminate side effects caused by the absence of nonlinear operations such as softmax for strict alignment. Furthermore, to improve the performance of MCCNet under complex light conditions, we introduce an illumination loss during training. Qualitative and quantitative evaluations demonstrate that MCCNet performs well in arbitrary video and image style transfer tasks. Code is available at https://github.com/kongxiuxiu/MCCNetV2.

In this paper, we introduce a new method for the task of interaction transfer. Given an example interaction between a source object and an agent, our method can automatically infer both surface and spatial relationships for the agent and target objects within the same category, yielding more accurate and valid transfers. Specifically, our method characterizes the example interaction using a combined spatial and surface representation. We correspond the agent points and object points related to the representation to the target object space using a learned spatial and surface correspondence field, which represents objects as deformed and rotated signed distance fields. With the corresponded points, an optimization is performed under the constraints of our spatial and surface interaction representation and additional regularization. Experiments conducted on human-chair and hand-mug interaction transfer tasks show that our approach can handle larger geometry and topology variations between source and target shapes, significantly outperforming state-of-the-art methods.

Sampling, grouping, and aggregation are three important components in the multi-scale analysis of point clouds. In this paper, we present a novel data-driven sampler learning strategy for point-wise analysis tasks. Unlike the widely used sampling technique, Farthest Point Sampling (FPS), we propose to learn sampling and downstream applications jointly. Our key insight is that uniform sampling methods like FPS are not always optimal for different tasks: sampling more points around boundary areas can make the point-wise classification easier for segmentation. Towards this end, we propose a novel sampler learning strategy that learns sampling point displacement supervised by task-related ground truth information and can be trained jointly with the underlying tasks. We further demonstrate our methods in various point-wise analysis tasks, including semantic part segmentation, point cloud completion, and keypoint detection. Our experiments show that jointly learning of the sampler and task brings better performance than using FPS in various point-based networks.

Given a 3D object, kinematic motion prediction aims to identify the mobile parts as well as the corresponding motion parameters. Due to the large variations in both topological structure and geometric details of 3D objects, this remains a challenging task and the lack of large scale labeled data also constrain the performance of deep learning based approaches. In this paper, we tackle the task of object kinematic motion prediction problem in a semi-weakly supervised manner. Our key observations are two-fold. First, although 3D dataset with fully annotated motion labels is limited, there are existing datasets and methods for object part semantic segmentation at large scale. Second, semantic part segmentation and mobile part segmentation is not always consistent but it is possible to detect the mobile parts from the underlying 3D structure. Towards this end, we propose a graph neural network to learn the map between hierarchical part-level segmentation and mobile parts parameters, which are further refined based on geometric alignment. This network can be first trained on PartNet-Mobility dataset with fully labeled mobility information and then applied on PartNet dataset with fine-grained and hierarchical part-level segmentation. The network predictions yield a large scale of 3D objects with pseudo labeled mobility information and can further be used for weakly-supervised learning with pre-existing segmentation. Our experiments show there are significant performance boosts with the augmented data for previous method designed for kinematic motion prediction on 3D partial scans.

The artistic style within a painting is the means of expression, which includes not only the painting material, colors, and brushstrokes, but also the high-level attributes, including semantic elements and object shapes. Previous arbitrary example-guided artistic image generation methods often fail to control shape changes or convey elements. Pre-trained text-to-image synthesis diffusion probabilistic models have achieved remarkable quality but often require extensive textual descriptions to accurately portray the attributes of a particular painting. The uniqueness of an artwork lies in the fact that it cannot be adequately explained with normal language. Our key idea is to learn the artistic style directly from a single painting and then guide the synthesis without providing complex textual descriptions. Specifically, we perceive style as a learnable textual description of a painting. We propose an inversion-based style transfer method (InST), which can efficiently and accurately learn the key information of an image, thus capturing and transferring the artistic style of a painting. We demonstrate the quality and efficiency of our method on numerous paintings of various artists and styles. Codes are available at https://github.com/zyxElsa/InST.

In this work, we tackle the challenging problem of learning-based single-view 3D hair modeling. Due to the great difficulty of collecting paired real image and 3D hair data, using synthetic data to provide prior knowledge for real domain becomes a leading solution. This unfortunately introduces the challenge of domain gap. Due to the inherent difficulty of realistic hair rendering, existing methods typically use orientation maps instead of hair images as input to bridge the gap. We firmly think an intermediate representation is essential, but we argue that orientation map using the dominant filtering-based methods is sensitive to uncertain noise and far from a competent representation. Thus, we first raise this issue up and propose a novel intermediate representation, termed as HairStep, which consists of a strand map and a depth map. It is found that HairStep not only provides sufficient information for accurate 3D hair modeling, but also is feasible to be inferred from real images. Specifically, we collect a dataset of 1,250 portrait images with two types of annotations. A learning framework is further designed to transfer real images to the strand map and depth map. It is noted that, an extra bonus of our new dataset is the first quantitative metric for 3D hair modeling. Our experiments show that HairStep narrows the domain gap between synthetic and real and achieves state-of-the-art performance on single-view 3D hair reconstruction.

Training generative adversarial networks (GANs) with limited data is challenging because the discriminator is prone to overfitting. Previously proposed differentiable augmentation demonstrates improved data efficiency of training GANs. However, the augmentation implicitly introduces undesired invariance to augmentation for the discriminator since it ignores the change of semantics in the label space caused by data transformation, which may limit the representation learning ability of the discriminator and ultimately affect the generative modeling performance of the generator. To mitigate the negative impact of invariance while inheriting the benefits of data augmentation, we propose a novel augmentation-aware self-supervised discriminator that predicts the augmentation parameter of the augmented data. Particularly, the prediction targets of real data and generated data are required to be distinguished since they are different during training. We further encourage the generator to adversarially learn from the self-supervised discriminator by generating augmentation-predictable real and not fake data. This formulation connects the learning objective of the generator and the arithmetic $-$ harmonic mean divergence under certain assumptions. We compare our method with state-of-the-art (SOTA) methods using the class-conditional BigGAN and unconditional StyleGAN2 architectures on data-limited CIFAR-10, CIFAR-100, FFHQ, LSUN-Cat, and five low-shot datasets. Experimental results demonstrate significant improvements of our method over SOTA methods in training data-efficient GANs.

In this paper, we propose a novel framework, Tracking-free Relightable Avatar (TRAvatar), for capturing and reconstructing high-fidelity 3D avatars. Compared to previous methods, TRAvatar works in a more practical and efficient setting. Specifically, TRAvatar is trained with dynamic image sequences captured in a Light Stage under varying lighting conditions, enabling realistic relighting and real-time animation for avatars in diverse scenes. Additionally, TRAvatar allows for tracking-free avatar capture and obviates the need for accurate surface tracking under varying illumination conditions. Our contributions are two-fold: First, we propose a novel network architecture that explicitly builds on and ensures the satisfaction of the linear nature of lighting. Trained on simple group light captures, TRAvatar can predict the appearance in real-time with a single forward pass, achieving high-quality relighting effects under illuminations of arbitrary environment maps. Second, we jointly optimize the facial geometry and relightable appearance from scratch based on image sequences, where the tracking is implicitly learned. This tracking-free approach brings robustness for establishing temporal correspondences between frames under different lighting conditions. Extensive qualitative and quantitative experiments demonstrate that our framework achieves superior performance for photorealistic avatar animation and relighting.

Abstract In this work, we introduce a new approach for face stylization. Despite existing methods achieving impressive results in this task, there is still room for improvement in generating high‐quality artistic faces with diverse styles and accurate facial reconstruction. Our proposed framework, MMFS, supports multi‐modal face stylization by leveraging the strengths of StyleGAN and integrates it into an encoder‐decoder architecture. Specifically, we use the mid‐resolution and high‐resolution layers of StyleGAN as the decoder to generate high‐quality faces, while aligning its low‐resolution layer with the encoder to extract and preserve input facial details. We also introduce a two‐stage training strategy, where we train the encoder in the first stage to align the feature maps with StyleGAN and enable a faithful reconstruction of input faces. In the second stage, the entire network is fine‐tuned with artistic data for stylized face generation. To enable the fine‐tuned model to be applied in zero‐shot and one‐shot stylization tasks, we train an additional mapping network from the large‐scale Contrastive‐Language‐Image‐Pre‐training (CLIP) space to a latent w+ space of fine‐tuned StyleGAN. Qualitative and quantitative experiments show that our framework achieves superior performance in both one‐shot and zero‐shot face stylization tasks, outperforming state‐of‐the‐art methods by a large margin.

Abstract In this paper, we tackle the challenging problem of 3D keypoint estimation of general objects using a novel implicit representation. Previous works have demonstrated promising results for keypoint prediction through direct coordinate regression or heatmap‐based inference. However, these methods are commonly studied for specific subjects, such as human bodies and faces, which possess fixed keypoint structures. They also suffer in several practical scenarios where explicit or complete geometry is not given, including images and partial point clouds. Inspired by the recent success of advanced implicit representation in reconstruction tasks, we explore the idea of using an implicit field to represent keypoints. Specifically, our key idea is employing spheres to represent 3D keypoints, thereby enabling the learnability of the corresponding signed distance field. Explicit key‐points can be extracted subsequently by our algorithm based on the Hough transform. Quantitative and qualitative evaluations also show the superiority of our representation in terms of prediction accuracy.

In this paper, we tackle the challenging problem of point cloud completion from the perspective of feature learning. Our key observation is that to recover the underlying structures as well as surface details, given partial input, a fundamental component is a good feature representation that can capture both global structure and local geometric details. We accordingly first propose FSNet, a feature structuring module that can adaptively aggregate point-wise features into a 2D structured feature map by learning multiple latent patterns from local regions. We then integrate FSNet into a coarse-to-fine pipeline for point cloud completion. Specifically, a 2D convolutional neural network is adopted to decode feature maps from FSNet into a coarse and complete point cloud. Next, a point cloud upsampling network is used to generate a dense point cloud from the partial input and the coarse intermediate output. To efficiently exploit local structures and enhance point distribution uniformity, we propose IFNet, a point upsampling module with a self-correction mechanism that can progressively refine details of the generated dense point cloud. We have conducted qualitative and quantitative experiments on ShapeNet, MVP, and KITTI datasets, which demonstrate that our method outperforms state-of-the-art point cloud completion approaches.

Personalizing generative models offers a way to guide image generation with user-provided references. Current personalization methods can invert an object or concept into the textual conditioning space and compose new natural sentences for text-to-image diffusion models. However, representing and editing specific visual attributes such as material, style, and layout remains a challenge, leading to a lack of disentanglement and editability. To address this problem, we propose a novel approach that leverages the step-by-step generation process of diffusion models, which generate images from low to high frequency information, providing a new perspective on representing, generating, and editing images. We develop the Prompt Spectrum Space P*, an expanded textual conditioning space, and a new image representation method called ProSpect. ProSpect represents an image as a collection of inverted textual token embeddings encoded from per-stage prompts, where each prompt corresponds to a specific generation stage (i.e., a group of consecutive steps) of the diffusion model. Experimental results demonstrate that P* and ProSpect offer better disentanglement and controllability compared to existing methods. We apply ProSpect in various personalized attribute-aware image generation applications, such as image-guided or text-driven manipulations of materials, style, and layout, achieving previously unattainable results from a single image input without fine-tuning the diffusion models. Our source code is available at https://github.com/zyxElsa/ProSpect.

This work presents Unified Contrastive Arbitrary Style Transfer (UCAST), a novel style representation learning and transfer framework, that can fit in most existing arbitrary image style transfer models, such as CNN-based, ViT-based, and flow-based methods. As the key component in image style transfer tasks, a suitable style representation is essential to achieve satisfactory results. Existing approaches based on deep neural networks typically use second-order statistics to generate the output. However, these hand-crafted features computed from a single image cannot leverage style information sufficiently, which leads to artifacts such as local distortions and style inconsistency. To address these issues, we learn style representation directly from a large number of images based on contrastive learning by considering the relationships between specific styles and the holistic style distribution. Specifically, we present an adaptive contrastive learning scheme for style transfer by introducing an input-dependent temperature. Our framework consists of three key components: a parallel contrastive learning scheme for style representation and transfer, a domain enhancement (DE) module for effective learning of style distribution, and a generative network for style transfer. Qualitative and quantitative evaluations show the results of our approach are superior to those obtained via state-of-the-art methods. The code is available at https://github.com/zyxElsa/CAST_pytorch .

In this work, we propose a framework for singleview hand mesh reconstruction, which can simultaneously achieve high reconstruction accuracy, fast inference speed, and temporal coherence. Specifically, for 2D encoding, we propose lightweight yet effective stacked structures. Regarding 3D decoding, we provide an efficient graph operator, namely depth-separable spiral convolution. Moreover, we present a novel feature lifting module for bridging the gap between 2D and 3D representations. This module begins with a map-based position regression (MapReg) block to integrate the merits of both heatmap encoding and position regression paradigms for improved 2D accuracy and temporal coherence. Furthermore, MapReg is followed by pose pooling and pose-to-vertex lifting approaches, which transform 2D pose encodings to semantic features of 3D vertices. Overall, our hand reconstruction framework, called MobRecon, comprises affordable computational costs and miniature model size, which reaches a high inference speed of 83FPS on Apple A14 CPU. Extensive experiments on popular datasets such as FreiHAND, RHD, and HO3Dv2 demonstrate that our MobRecon achieves superior performance on reconstruction accuracy and temporal coherence. Our code is publicly available at https://github.com/SeanChenxy/HandMesh.

The goal of image style transfer is to render an image with artistic features guided by a style reference while maintaining the original content. Owing to the locality in convolutional neural networks (CNNs), extracting and maintaining the global information of input images is difficult. Therefore, traditional neural style transfer methods face biased content representation. To address this critical issue, we take long-range dependencies of input images into account for image style transfer by proposing a transformer-based approach called StyTr2. In contrast with visual transformers for other vision tasks, StyTr2 contains two different transformer encoders to generate domain-specific sequences for content and style, respectively. Following the encoders, a multi-layer transformer decoder is adopted to stylize the content sequence according to the style sequence. We also analyze the deficiency of existing positional encoding methods and propose the content-aware positional encoding (CAPE), which is scale-invariant and more suitable for image style transfer tasks. Qualitative and quantitative experiments demonstrate the effectiveness of the proposed StyTr2 compared with state-of-the-art CNN-based and flow-based approaches. Code and models are available at https://github.com/diyiiyiii/StyTR-2.

Abstract In this work, we present a new method for 3D face reconstruction from sparse‐view RGB images. Unlike previous methods which are built upon 3D morphable models (3DMMs) with limited details, we leverage an implicit representation to encode rich geometric features. Our overall pipeline consists of two major components, including a geometry network, which learns a deformable neural signed distance function (SDF) as the 3D face representation, and a rendering network, which learns to render on‐surface points of the neural SDF to match the input images via self‐supervised optimization. To handle in‐the‐wild sparse‐view input of the same target with different expressions at test time, we propose residual latent code to effectively expand the shape space of the learned implicit face representation as well as a novel view‐switch loss to enforce consistency among different views. Our experimental results on several benchmark datasets demonstrate that our approach outperforms alternative baselines and achieves superior face reconstruction results compared to state‐of‐the‐art methods.

Quantization has proven effective in high-resolution and large-scale simulations, which benefit from bit-level memory saving. However, identifying a quantization scheme that meets the requirement of both precision and memory efficiency requires trial and error. In this paper, we propose a novel framework to allow users to obtain a quantization scheme by simply specifying either an error bound or a memory compression rate. Based on the error propagation theory, our method takes advantage of auto-diff to estimate the contributions of each quantization operation to the total error. We formulate the task as a constrained optimization problem, which can be efficiently solved with analytical formulas derived for the linearized objective function. Our workflow extends the Taichi compiler and introduces dithering to improve the precision of quantized simulations. We demonstrate the generality and efficiency of our method via several challenging examples of physics-based simulation, which achieves up to 2.5× memory compression without noticeable degradation of visual quality in the results. Our code and data are available at https://github.com/Hanke98/AutoQantizer.

Video style transfer is attracting increasing attention from the artificial intelligence community because of its numerous applications, such as augmented reality and animation production. Relative to traditional image style transfer, video style transfer presents new challenges, including how to effectively generate satisfactory stylized results for any specified style while maintaining temporal coherence across frames. Towards this end, we propose a Multi-Channel Correlation network (MCCNet), which can be trained to fuse exemplar style features and input content features for efficient style transfer while naturally maintaining the coherence of input videos to output videos. Specifically, MCCNet works directly on the feature space of style and content domain where it learns to rearrange and fuse style features on the basis of their similarity to content features. The outputs generated by MCC are features containing the desired style patterns that can further be decoded into images with vivid style textures. Moreover, MCCNet is also designed to explicitly align the features to input and thereby ensure that the outputs maintain the content structures and the temporal continuity. To further improve the performance of MCCNet under complex light conditions, we also introduce illumination loss during training. Qualitative and quantitative evaluations demonstrate that MCCNet performs well in arbitrary video and image style transfer tasks. Code is available at https://github.com/diyiiyiii/MCCNet.

Recent years have witnessed significant progress in 3D hand mesh recovery. Nevertheless, because of the intrinsic 2D-to-3D ambiguity, recovering camera-space 3D information from a single RGB image remains challenging. To tackle this problem, we divide camera-space mesh recovery into two sub-tasks, i.e., root-relative mesh recovery and root recovery. First, joint landmarks and silhouette are extracted from a single input image to provide 2D cues for the 3D tasks. In the root-relative mesh recovery task, we exploit semantic relations among joints to generate a 3D mesh from the extracted 2D cues. Such generated 3D mesh coordinates are expressed relative to a root position, i.e., wrist of the hand. In the root recovery task, the root position is registered to the camera space by aligning the generated 3D mesh back to 2D cues, thereby completing camera-space 3D mesh recovery. Our pipeline is novel in that (1) it explicitly makes use of known semantic relations among joints and (2) it exploits 1D projections of the silhouette and mesh to achieve robust registration. Extensive experiments on popular datasets such as FreiHAND, RHD, and Human3.6M demonstrate that our approach achieves state-of-the-art performance on both root-relative mesh recovery and root recovery. Our code is publicly available at https://github.com/SeanChenxy/HandMesh.

Developing deep neural networks to generate 3D scenes is a fundamental problem in neural synthesis with immediate applications in architectural CAD, computer graphics, as well as in generating virtual robot training environments. This task is challenging because 3D scenes exhibit diverse patterns, ranging from continuous ones, such as object sizes and the relative poses between pairs of shapes, to discrete patterns, such as occurrence and co-occurrence of objects with symmetrical relationships. This paper introduces a novel neural scene synthesis approach that can capture diverse feature patterns of 3D scenes. Our method combines the strength of both neural network-based and conventional scene synthesis approaches. We use the parametric prior distributions learned from training data, which provide uncertainties of object attributes and relative attributes, to regularize the outputs of feed-forward neural models. Moreover, instead of merely predicting a scene layout, our approach predicts an over-complete set of attributes. This methodology allows us to utilize the underlying consistency constraints among the predicted attributes to prune infeasible predictions. Experimental results show that our approach outperforms existing methods considerably. The generated 3D scenes interpolate the training data faithfully while preserving both continuous and discrete feature patterns.

This paper introduces HPNet, a novel deep-learning approach for segmenting a 3D shape represented as a point cloud into primitive patches. The key to deep primitive segmentation is learning a feature representation that can separate points of different primitives. Unlike utilizing a single feature representation, HPNet leverages hybrid representations that combine one learned semantic descriptor, two spectral descriptors derived from predicted geometric parameters, as well as an adjacency matrix that encodes sharp edges. Moreover, instead of merely concatenating the descriptors, HPNet optimally combines hybrid representations by learning combination weights. This weighting module builds on the entropy of input features. The output primitive segmentation is obtained from a meanshift clustering module. Experimental results on benchmark datasets ANSI and ABCParts show that HPNet leads to significant performance gains from baseline approaches.

Distinguishing aesthetically pleasing food photos from others is an important visual analysis task for social media and ranking systems related to food. Nevertheless, aesthetic assessment of food images remains a challenging and relatively unexplored task, largely due to the lack of related food image datasets and practical knowledge. Thus, we present the Gourmet Photography Dataset (GPD), the first large-scale dataset for aesthetic assessment of food photos. It contains 24,000 images with corresponding binary aesthetic labels, covering a large variety of foods and scenes. We also provide a non-stationary regularization method to combat over-fitting and enhance the ability of tuned models to generalize. Quantitative results from extensive experiments, including a generalization ability test, verify that neural networks trained on the GPD achieve comparable performance to human experts on the task of aesthetic assessment. We reveal several valuable findings to support further research and applications related to visual aesthetic analysis of food images. To encourage further research, we have made the GPD publicly available at https://github.com/Openning07/GPA .

Art painting evaluation is sophisticated for a novice with no or limited knowledge on art criticism, and history. In this study, we propose the concept ofrepresentativityto evaluate paintings instead of using professional concepts, such as genre, media, and style, which may be confusing to non-professionals. We define the concept of representativity to evaluate quantitatively the extent to which a painting can represent the characteristics of an artists creations. We begin by proposing a novel deep representation of art paintings, which is enhanced by style information through a weighted pooling feature fusion module. In contrast to existing feature extraction approaches, the proposed framework embeds painting styles, and authorship information, and learns specific artwork characteristics in a single framework. Subsequently, we propose a graph-based learning method for representativity learning, which considers intra-category, and extra-category information. In view of the significance of historical factors in the art domain, we introduce the creation time of a painting into the learning process. User studies demonstrate our approach helps the public effectively access the creation characteristics of artists through sorting paintings by representativity from highest to lowest.

Multimodal and multi-domain stylization are two important problems in the field of image style transfer. Currently, there are few methods that can perform multimodal and multi-domain stylization simultaneously. In this study, we propose a unified framework for multimodal and multi-domain style transfer with the support of both exemplar-based reference and randomly sampled guidance. The key component of our method is a novel style distribution alignment module that eliminates the explicit distribution gaps between various style domains and reduces the risk of mode collapse. The multimodal diversity is ensured by either guidance from multiple images or random style codes, while the multi-domain controllability is directly achieved by using a domain label. We validate our proposed framework on painting style transfer with various artistic styles and genres. Qualitative and quantitative comparisons with state-of-the-art methods demonstrate that our method can generate high-quality results of multi-domain styles and multimodal instances from reference style guidance or a random sampled style.

With the surge of images in the information era, people demand an effective and accurate way to access meaningful visual information. Accordingly, effective and accurate communication of information has become indispensable. In this article, we propose a content-based approach that automatically generates a clear and informative visual summarization based on design principles and cognitive psychology to represent image collections. We first introduce a novel method to make representative and nonredundant summarizations of image collections, thereby ensuring data cleanliness and emphasizing important information. Then, we propose a tree-based algorithm with a two-step optimization strategy to generate the final layout that operates as follows: (1) an initial layout is created by constructing a tree randomly based on the grouping results of the input image set; (2) the layout is refined through a coarse adjustment in a greedy manner, followed by gradient back propagation drawing on the training procedure of neural networks. We demonstrate the usefulness and effectiveness of our method via extensive experimental results and user studies. Our visual summarization algorithm can precisely and efficiently capture the main content of image collections better than alternative methods or commercial tools.

Visual aesthetic assessment has been an active research field for decades. Although latest methods have achieved promising performance on benchmark datasets, they typically rely on a large number of manual annotations including both aesthetic labels and related image attributes. In this paper, we revisit the problem of image aesthetic assessment from the self-supervised feature learning perspective. Our motivation is that a suitable feature representation for image aesthetic assessment should be able to distinguish different expert-designed image manipulations, which have close relationships with negative aesthetic effects. To this end, we design two novel pretext tasks to identify the types and parameters of editing operations applied to synthetic instances. The features from our pretext tasks are then adapted for a one-layer linear classifier to evaluate the performance in terms of binary aesthetic classification. We conduct extensive quantitative experiments on three benchmark datasets and demonstrate that our approach can faithfully extract aesthetics-aware features and outperform alternative pretext schemes. Moreover, we achieve comparable results to state-of-the-art supervised methods that use 10 million labels from ImageNet.

Object detection has achieved remarkable progress in the past decade. However, the detection of oriented and densely packed objects remains challenging because of following inherent reasons: (1) receptive fields of neurons are all axis-aligned and of the same shape, whereas objects are usually of diverse shapes and align along various directions; (2) detection models are typically trained with generic knowledge and may not generalize well to handle specific objects at test time; (3) the limited dataset hinders the development on this task. To resolve the first two issues, we present a dynamic refinement network that consists of two novel components, i.e., a feature selection module (FSM) and a dynamic refinement head (DRH). Our FSM enables neurons to adjust receptive fields in accordance with the shapes and orientations of target objects, whereas the DRH empowers our model to refine the prediction dynamically in an object-aware manner. To address the limited availability of related benchmarks, we collect an extensive and fully annotated dataset, namely, SKU110K-R, which is relabeled with oriented bounding boxes based on SKU110K. We perform quantitative evaluations on several publicly available benchmarks including DOTA, HRSC2016, SKU110K, and our own SKU110K-R dataset. Experimental results show that our method achieves consistent and substantial gains compared with baseline approaches. Our source code and dataset will be released to encourage follow-up research.

Real-world applications could benefit from the ability to automatically retarget an image to different aspect ratios and resolutions while preserving its visually and semantically important content. However, not all images can be equally processed. This study introduces the notion of image retargetability to describe how well a particular image can be handled by content-aware image retargeting. We propose to learn a deep convolutional neural network to rank photo retargetability, in which the relative ranking of photo retargetability is directly modeled in the loss function. Our model incorporates the joint learning of meaningful photographic attributes and image content information, which can facilitate the regularization of the complicated retargetability rating problem. To train and analyze this model, we collect a dataset that contains retargetability scores and meaningful image attributes assigned by six expert raters. The experiments demonstrate that our unified model can generate retargetability rankings that are highly consistent with human labels. To further validate our model, we show the applications of image retargetability in retargeting method selection, retargeting method assessment and generating a photo collage.

We present a deep generative scene modeling technique for indoor environments. Our goal is to train a generative model using a feed-forward neural network that maps a prior distribution (e.g., a normal distribution) to the distribution of primary objects in indoor scenes. We introduce a 3D object arrangement representation that models the locations and orientations of objects, based on their size and shape attributes. Moreover, our scene representation is applicable for 3D objects with different multiplicities (repetition counts), selected from a database. We show a principled way to train this model by combining discriminative losses for both a 3D object arrangement representation and a 2D image-based representation. We demonstrate the effectiveness of our scene representation and the network training method on benchmark datasets. We also show the applications of this generative model in scene interpolation and scene completion.

Time-lapse videos usually contain visually appealing content but are often difficult and costly to create. In this paper, we present an end-to-end solution to synthesize a time-lapse video from a single outdoor image using deep neural networks. Our key idea is to train a conditional generative adversarial network based on existing datasets of time-lapse videos and image sequences. We propose a multi-frame joint conditional generation framework to effectively learn the correlation between the illumination change of an outdoor scene and the time of the day. We further present a multi-domain training scheme for robust training of our generative models from two datasets with different distributions and missing timestamp labels. Compared to alternative time-lapse video synthesis algorithms, our method uses the timestamp as the control variable and does not require a reference video to guide the synthesis of the final output. We conduct ablation studies to validate our algorithm and compare with state-of-the-art techniques both qualitatively and quantitatively.

We introduce a new silhouette-based representation for modeling clothed human bodies using deep generative models. Our method can reconstruct a complete and textured 3D model of a person wearing clothes from a single input picture. Inspired by the visual hull algorithm, our implicit representation uses 2D silhouettes and 3D joints of a body pose to describe the immense shape complexity and variations of clothed people. Given a segmented 2D silhouette of a person and its inferred 3D joints from the input picture, we first synthesize consistent silhouettes from novel view points around the subject. The synthesized silhouettes which are the most consistent with the input segmentation are fed into a deep visual hull algorithm for robust 3D shape prediction. We then infer the texture of the subject's back view using the frontal image and segmentation mask as input to a conditional generative adversarial network. Our experiments demonstrate that our silhouette-based model is an effective representation and the appearance of the back view can be predicted reliably using an image-to-image translation network. While classic methods based on parametric models often fail for single-view images of subjects with challenging clothing, our approach can still produce successful results, which are comparable to those obtained from multi-view input.

Taking a satisfactory picture in a low-light environment remains a challenging problem. Low-light imaging mainly suffers from noise due to the low signal-to-noise ratio. Many methods have been proposed for the task of image denoising, but they fail to work under extremely low-light conditions. Recently, deep learning based approaches have been presented that have higher objective quality than traditional methods, but they usually have high computational cost which makes them impractical to use in real-time applications or where the processing power is limited. In this paper, we propose a new residual learning based deep neural network for end-to-end extreme low-light image denoising that can not only significantly reduce the computational cost but also improve the quality over existing methods in both objective and subjective metrics. Specifically, in one setting we achieved 29x speedup with higher PSNR. Subjectively, our method provides better color reproduction and preserves more detailed texture information compared to state-of-the-art methods.

In this work, we propose a novel meta-learning approach for few-shot classification, which learns transferable prior knowledge across tasks and directly produces network parameters for similar unseen tasks with training samples. Our approach, called LGM-Net, includes two key modules, namely, TargetNet and MetaNet. The TargetNet module is a neural network for solving a specific task and the MetaNet module aims at learning to generate functional weights for TargetNet by observing training samples. We also present an intertask normalization strategy for the training process to leverage common information shared across different tasks. The experimental results on Omniglot and miniImageNet datasets demonstrate that LGM-Net can effectively adapt to similar unseen tasks and achieve competitive performance, and the results on synthetic datasets show that transferable prior knowledge is learned by the MetaNet module via mapping training data to functional weights. LGM-Net enables fast learning and adaptation since no further tuning steps are required compared to other meta-learning approaches

User interaction provides useful information for solving challenging computer vision problems in practice. In this paper, we show that a very limited number of user clicks could greatly boost monocular depth estimation performance and overcome monocular ambiguities. We formulate this task as a deep structured model, in which the structured pixel-wise depth estimation has ordinal constraints introduced by user clicks. We show that the inference of the proposed model could be efficiently solved through a feed-forward network. We demonstrate the effectiveness of the proposed model on NYU Depth V2 and Stanford 2D-3D datasets. On both datasets, we achieve state-of-the-art performance when encoding user interaction into our deep models.

Aggregation structures with explicit information, such as image attributes and scene semantics, are effective and popular for intelligent systems for assessing aesthetics of visual data. However, useful information may not be available due to the high cost of manual annotation and expert design. In this paper, we present a novel multi-patch (MP) aggregation method for image aesthetic assessment. Different from state-of-the-art methods, which augment an MP aggregation network with various visual attributes, we train the model in an end-to-end manner with aesthetic labels only (i.e., aesthetically positive or negative). We achieve the goal by resorting to an attention-based mechanism that adaptively adjusts the weight of each patch during the training process to improve learning efficiency. In addition, we propose a set of objectives with three typical attention mechanisms (i.e., average, minimum, and adaptive) and evaluate their effectiveness on the Aesthetic Visual Analysis (AVA) benchmark. Numerical results show that our approach outperforms existing methods by a large margin. We further verify the effectiveness of the proposed attention-based objectives via ablation studies and shed light on the design of aesthetic assessment systems.

Recent advances in single-view 3D hair digitization have made the creation of high-quality CG characters scalable and accessible to end-users, enabling new forms of personalized VR and gaming experiences. To handle the complexity and variety of hair structures, most cutting-edge techniques rely on the successful retrieval of a particular hair model from a comprehensive hair database. Not only are the aforementioned data-driven methods storage intensive, but they are also prone to failure for highly unconstrained input images, complicated hairstyles, and failed face detection. Instead of using a large collection of 3D hair models directly, we propose to represent the manifold of 3D hairstyles implicitly through a compact latent space of a volumetric variational autoencoder (VAE). This deep neural network is trained with volumetric orientation field representations of 3D hair models and can synthesize new hairstyles from a compressed code. To enable end-to-end 3D hair inference, we train an additional embedding network to predict the code in the VAE latent space from any input image. Strand-level hairstyles can then be generated from the predicted volumetric representation. Our fully automatic framework does not require any ad-hoc face fitting, intermediate classification and segmentation, or hairstyle database retrieval. Our hair synthesis approach is significantly more robust and can handle a much wider variation of hairstyles than state-of-the-art data-driven hair modeling techniques with challenging inputs, including photos that are low-resolution, overexposured, or contain extreme head poses. The storage requirements are minimal and a 3D hair model can be produced from an image in a second. Our evaluations also show that successful reconstructions are possible from highly stylized cartoon images, non-human subjects, and pictures taken from behind a person. Our approach is particularly well suited for continuous and plausible hair interpolation between very different hairstyles.

The visual richness of computer graphics applications is frequently limited by the difficulty of obtaining high-quality, detailed 3D models. This paper proposes a method for realistically transferring details (specifically, displacement maps) from existing high-quality 3D models to simple shapes that may be created with easy-to-learn modeling tools. Our key insight is to use metric learning to find a combination of geometric features that successfully predicts detail-map similarities on the source mesh; we use the learned feature combination to drive the detail transfer. The latter uses a variant of multi-resolution non-parametric texture synthesis, augmented by a high-frequency detail transfer step in texture space. We demonstrate that our technique can successfully transfer details among a variety of shapes including furniture and clothing.

This paper presents a data-driven approach for automatically generating cartoon faces in different styles from a given portrait image. Our stylization pipeline consists of two steps: an offline analysis step to learn about how to select and compose facial components from the databases; a runtime synthesis step to generate the cartoon face by assembling parts from a database of stylized facial components. We propose an optimization framework that, for a given artistic style, simultaneously considers the desired image-cartoon relationships of the facial components and a proper adjustment of the image composition. We measure the similarity between facial components of the input image and our cartoon database via image feature matching, and introduce a probabilistic framework for modeling the relationships between cartoon facial components. We incorporate prior knowledge about image-cartoon relationships and the optimal composition of facial components extracted from a set of cartoon faces to maintain a natural, consistent, and attractive look of the results. We demonstrate generality and robustness of our approach by applying it to a variety of portrait images and compare our output with stylized results created by artists via a comprehensive user study.

3D tensor field design is important in several graphics applications such as procedural noise, solid texturing, and geometry synthesis. Different fields can lead to different visual effects. The topology of a tensor field, such as degenerate tensors, can cause artifacts in these applications. Existing 2D tensor field design systems cannot be used to handle the topology of a 3D tensor field. In this paper, we present to our knowledge the first 3D tensor field design system. At the core of our system is the ability to edit the topology of tensor fields. We demonstrate the power of our design system with applications in solid texturing and geometry synthesis.

Similar objects are ubiquitous and abundant in both natural and artificial scenes. Determining the visual importance of several similar objects in a complex photograph is a challenge for image understanding algorithms. This study aims to define the importance of similar objects in an image and to develop a method that can select the most important instances for an input image from multiple similar objects. This task is challenging because multiple objects must be compared without adequate semantic information. This challenge is addressed by building an image database and designing an interactive system to measure object importance from human observers. This ground truth is used to define a range of features related to the visual importance of similar objects. Then, these features are used in learning-to-rank and random forest to rank similar objects in an image. Importance predictions were validated on 5,922 objects. The most important objects can be identified automatically. The factors related to composition (e.g., size, location, and overlap) are particularly informative, although clarity and color contrast are also important. We demonstrate the usefulness of similar object importance on various applications, including image retargeting, image compression, image re-attentionizing, image admixture, and manipulation of blindness images.

We introduce a realtime facial tracking system specifically designed for performance capture in unconstrained settings using a consumer-level RGB-D sensor. Our framework provides uninterrupted 3D facial tracking, even in the presence of extreme occlusions such as those caused by hair, hand-to-face gestures, and wearable accessories. Anyone's face can be instantly tracked and the users can be switched without an extra calibration step. During tracking, we explicitly segment face regions from any occluding parts by detecting outliers in the shape and appearance input using an exponentially smoothed and user-adaptive tracking model as prior. Our face segmentation combines depth and RGB input data and is also robust against illumination changes. To enable continuous and reliable facial feature tracking in the color channels, we synthesize plausible face textures in the occluded regions. Our tracking model is personalized on-the-fly by progressively refining the user's identity, expressions, and texture with reliable samples and temporal filtering. We demonstrate robust and high-fidelity facial tracking on a wide range of subjects with highly incomplete and largely occluded data. Our system works in everyday environments and is fully unobtrusive to the user, impacting consumer AR applications and surveillance.

Human hair presents highly convoluted structures and spans an extraordinarily wide range of hairstyles, which is essential for the digitization of compelling virtual avatars but also one of the most challenging to create. Cutting-edge hair modeling techniques typically rely on expensive capture devices and significant manual labor. We introduce a novel data-driven framework that can digitize complete and highly complex 3D hairstyles from a single-view photograph. We first construct a large database of manually crafted hair models from several online repositories. Given a reference photo of the target hairstyle and a few user strokes as guidance, we automatically search for multiple best matching examples from the database and combine them consistently into a single hairstyle to form the large-scale structure of the hair model. We then synthesize the final hair strands by jointly optimizing for the projected 2D similarity to the reference photo, the physical plausibility of each strand, as well as the local orientation coherency between neighboring strands. We demonstrate the effectiveness and robustness of our method on a variety of hairstyles and challenging images, and compare our system with state-of-the-art hair modeling algorithms.

There are currently no solutions for enabling direct face-to-face interaction between virtual reality (VR) users wearing head-mounted displays (HMDs). The main challenge is that the headset obstructs a significant portion of a user's face, preventing effective facial capture with traditional techniques. To advance virtual reality as a next-generation communication platform, we develop a novel HMD that enables 3D facial performance-driven animation in real-time. Our wearable system uses ultra-thin flexible electronic materials that are mounted on the foam liner of the headset to measure surface strain signals corresponding to upper face expressions. These strain signals are combined with a head-mounted RGB-D camera to enhance the tracking in the mouth region and to account for inaccurate HMD placement. To map the input signals to a 3D face model, we perform a single-instance offline training session for each person. For reusable and accurate online operation, we propose a short calibration step to readjust the Gaussian mixture distribution of the mapping before each use. The resulting animations are visually on par with cutting-edge depth sensor-driven facial performance capture systems and hence, are suitable for social interactions in virtual worlds.

Abstract Style transfer aims to apply the style of an exemplar model to a target one, while retaining the target's structure. The main challenge in this process is to algorithmically distinguish style from structure, a high‐level, potentially ill‐posed cognitive task. Inspired by cognitive science research we recast style transfer in terms ofshape analogies. In IQ testing, shape analogy queries present the subject with three shapes:source, targetandexemplar, and ask them to select anoutputsuch that the transformation, oranalogy, from the exemplar to the output is similar to that from the source to the target. The logical process involved in identifying the source‐to‐target analogies implicitly detects the structural differences between the source and target and can be used effectively to facilitate style transfer. Since the exemplar has a similar structure to the source, applying the analogy to the exemplar will provide the output we seek. The main technical challenge we address is to compute the source to target analogies, consistent with human logic. We observe that the typical analogies we look for consist of a small set of simple transformations, which when applied to the exemplar generate a continuous, seamless output model. To assemble a shape analogy, we compute an optimal set of source‐to‐target transformations, such that the assembled analogy best fits these criteria. The assembled analogy is then applied to the exemplar shape to produce the desired output model. We use the proposed framework to seamlessly transfer a variety of style properties between 2D and 3D objects and demonstrate significant improvements over the state of the art in style transfer. We further show that our framework can be used to successfully complete partial scans with the help of a user provided structural template, coherently propagating scan style across the completed surfaces.

Abstract The design of video game environments, or levels, aims to control gameplay by steering the player through a sequence of designer‐controlled steps, while simultaneously providing a visually engaging experience. Traditionally these levels are painstakingly designed by hand, often from pre‐existing building blocks, or space templates. In this paper, we propose an algorithmic approach for automatically laying out game levels from user‐specified blocks. Our method allows designers to retain control of the gameplay flow via user‐specified level connectivity graphs, while relieving them from the tedious task of manually assembling the building blocks into a valid, plausible layout. Our method produces sequences of diverse layouts for the same input connectivity, allowing for repeated replay of a given level within a visually different, new environment. We support complex graph connectivities and various building block shapes, and are able to compute complex layouts in seconds. The two key components of our algorithm are the use of configuration spaces defining feasible relative positions of building blocks within a layout and a graph‐decomposition based layout strategy that leverages graph connectivity to speed up convergence and avoid local minima. Together these two tools quickly steer the solution toward feasible layouts. We demonstrate our method on a variety of real‐life inputs, and generate appealing layouts conforming to user specifications.

We introduce a data-driven hair capture framework based on example strands generated through hair simulation. Our method can robustly reconstruct faithful 3D hair models from unprocessed input point clouds with large amounts of outliers. Current state-of-the-art techniques use geometrically-inspired heuristics to derive global hair strand structures, which can yield implausible hair strands for hairstyles involving large occlusions, multiple layers, or wisps of varying lengths. We address this problem using a voting-based fitting algorithm to discover structurally plausible configurations among the locally grown hair segments from a database of simulated examples. To generate these examples, we exhaustively sample the simulation configurations within the feasible parameter space constrained by the current input hairstyle. The number of necessary simulations can be further reduced by leveraging symmetry and constrained initial conditions. The final hairstyle can then be structurally represented by a limited number of examples. To handle constrained hairstyles such as a ponytail of which realistic simulations are more difficult, we allow the user to sketch a few strokes to generate strand examples through an intuitive interface. Our approach focuses on robustness and generality. Since our method is structurally plausible by construction, we ensure an improved control during hair digitization and avoid implausible hair synthesis for a wide range of hairstyles.

From fishtail to princess braids, these intricately woven structures define an important and popular class of hairstyle, frequently used for digital characters in computer graphics. In addition to the challenges created by the infinite range of styles, existing modeling and capture techniques are particularly constrained by the geometric and topological complexities. We propose a data-driven method to automatically reconstruct braided hairstyles from input data obtained from a single consumer RGB-D camera. Our approach covers the large variation of repetitive braid structures using a family of compact procedural braid models. From these models, we produce a database of braid patches and use a robust random sampling approach for data fitting. We then recover the input braid structures using a multi-label optimization algorithm and synthesize the intertwining hair strands of the braids. We demonstrate that a minimal capture equipment is sufficient to effectively capture a wide range of complex braids with distinct shapes and structures.

Many natural phenomena consist of geometric elements with dynamic motions characterized by small scale repetitions over large scale structures, such as particles, herds, threads, and sheets. Due to their ubiquity, controlling the appearance and behavior of such phenomena is important for a variety of graphics applications. However, such control is often challenging; the repetitive elements are often too numerous for manual edit, while their overall structures are often too versatile for fully automatic computation. We propose a method that facilitates easy and intuitive controls at both scales: high-level structures through spatial-temporal output constraints (e.g. overall shape and motion of the output domain), and low-level details through small input exemplars (e.g. element arrangements and movements). These controls are suitable for manual specification, while the corresponding geometric and dynamic repetitions are suitable for automatic computation. Our system takes such user controls as inputs, and generates as outputs the corresponding repetitions satisfying the controls. Our method, which we call dynamic element textures , aims to produce such controllable repetitions through a combination of constrained optimization (satisfying controls) and data driven computation (synthesizing details). We use spatial-temporal samples as the core representation for dynamic geometric elements. We propose analysis algorithms for decomposing small scale repetitions from large scale themes, as well as synthesis algorithms for generating outputs satisfying user controls. Our method is general, producing a range of artistic effects that previously required disparate and specialized techniques.

Abstract Beam tracing combines the flexibility of ray tracing and the speed of polygon rasterization. However, beam tracing so far only handles linear transformations; thus, it is only applicable to linear effects such as planar mirror reflections but not to non‐linear effects such as curved mirror reflection, refraction, caustics and shadows. In this paper, we introduce non‐linear beam tracing to render these non‐linear effects. Non‐linear beam tracing is highly challenging because commodity graphics hardware supports only linear vertex transformation and triangle rasterization. We overcome this difficulty by designing a non‐linear graphics pipeline and implementing it on top of a commodity GPU. This allows beams to be non‐linear where rays within the same beam do not have to be parallel or intersect at a single point. Using these non‐linear beams, real‐time GPU applications can render secondary rays via polygon streaming similar to how they render primary rays. A major strength of this methodology is that it naturally supports fully dynamic scenes without the need to pre‐store a scene database. Utilizing our approach, non‐linear ray tracing effects can be rendered in real‐time on a commodity GPU under a unified framework.

A variety of phenomena can be characterized by repetitive small scale elements within a large scale domain. Examples include a stack of fresh produce, a plate of spaghetti, or a mosaic pattern. Although certain results can be produced via manual placement or procedural/physical simulation, these methods can be labor intensive, difficult to control, or limited to specific phenomena. We present discrete element textures, a data-driven method for synthesizing repetitive elements according to a small input exemplar and a large output domain. Our method preserves both individual element properties and their aggregate distributions. It is also general and applicable to a variety of phenomena, including different dimensionalities, different element properties and distributions, and different effects including both artistic and physically realistic ones. We represent each element by one or multiple samples whose positions encode relevant element attributes including position, size, shape, and orientation. We propose a sample-based neighborhood similarity metric and an energy optimization solver to synthesize desired outputs that observe not only input exemplars and output domains but also optional constraints such as physics, orientation fields, and boundary conditions. As a further benefit, our method can also be applied for editing existing element distributions.

A variety of animation effects such as herds and fluids contain detailed motion fields characterized by repetitive structures. Such detailed motion fields are often visually important, but tedious to specify manually or expensive to simulate computationally. Due to the repetitive nature, some of these motion fields (e.g. turbulence in fluids) could be synthesized by procedural texturing, but procedural texturing is known for its limited generality. We apply example-based texture synthesis for motion fields. Our technique is general and can take on a variety of user inputs, including captured data, manual art, and physical/procedural simulation. This data-driven approach enables artistic effects that are difficult to achieve via previous methods, such as heart shaped swirls in fluid animation. Due to the use of texture synthesis, our method is able to populate a large output field from a small input exemplar, imposing minimum user workload. Our algorithm also allows the synthesis of output motion fields not only with the same dimension as the input (e.g. 2D to 2D) but also of higher dimension, such as 3D volumetric outputs from 2D planar inputs. This cross-dimension capability supports a convenient usage scenario, i.e. the user could simply supply 2D images and our method produces a 3D motion field with similar characteristics. The motion fields produced by our method are generic, and could be combined with a variety of large-scale low-resolution motions that are easy to specify either manually or computationally but lack the repetitive structures to be characterized as textures. We apply our technique to a variety of animation phenomena, including smoke, liquid, and group motion.