Changqing Zou

Ye Wang, Bowei Jiang, Changqing Zou et al.

Multifold observations are common for different data modalities, e.g., a 3D shape can be represented by multi-view images and an image can be described with different captions. Existing cross-modal contrastive representation learning (XM-CLR) methods such as CLIP are not fully suitable for multifold data as they only consider one positive pair and treat other pairs as negative when computing the contrastive loss. In this paper, we propose MXM-CLR, a unified framework for contrastive learning of multifold cross-modal representations. MXM-CLR explicitly models and learns the relationships between multifold observations of instances from different modalities for more comprehensive representation learning. The key of MXM-CLR is a novel multifold-aware hybrid loss which considers multiple positive observations when computing the hard and soft relationships for the cross-modal data pairs. We conduct quantitative and qualitative comparisons with SOTA baselines for cross-modal retrieval tasks on the Text2Shape and Flickr30K datasets. We also perform extensive evaluations on the adaptability and generalizability of MXM-CLR, as well as ablation studies on the loss design and effects of batch sizes. The results show the superiority of MXM-CLR in learning better representations for the multifold data. The code is available at https://github.com/JLU-ICL/MXM-CLR.

Mingrui Zhao, Yizhi Wang, Fenggen Yu et al.

Shape abstraction is an important task for simplifying complex geometric structures while retaining essential features. Sweep surfaces, commonly found in human-made objects, aid in this process by effectively capturing and representing object geometry, thereby facilitating abstraction. In this paper, we introduce \papername, a novel approach to shape abstraction through sweep surfaces. We propose an effective parameterization for sweep surfaces, utilizing superellipses for profile representation and B-spline curves for the axis. This compact representation, requiring as few as 14 float numbers, facilitates intuitive and interactive editing while preserving shape details effectively. Additionally, by introducing a differentiable neural sweeper and an encoder-decoder architecture, we demonstrate the ability to predict sweep surface representations without supervision. We show the superiority of our model through several quantitative and qualitative experiments throughout the paper. Our code is available at https://mingrui-zhao.github.io/SweepNet/

Yunpeng Qing, Yixiao Chi, Shuo Chen et al.

Recent advances in offline Reinforcement Learning (RL) have proven that effective policy learning can benefit from imposing conservative constraints on pre-collected datasets. However, such static datasets often exhibit distribution bias, resulting in limited generalizability. To address this limitation, a straightforward solution is data augmentation (DA), which leverages generative models to enrich data distribution. Despite the promising results, current DA techniques focus solely on reconstructing future trajectories from given states, while ignoring the exploration of history transitions that reach them. This single-direction paradigm inevitably hinders the discovery of diverse behavior patterns, especially those leading to critical states that may have yielded high-reward outcomes. In this work, we introduce Bidirectional Trajectory Diffusion (BiTrajDiff), a novel DA framework for offline RL that models both future and history trajectories from any intermediate states. Specifically, we decompose the trajectory generation task into two independent yet complementary diffusion processes: one generating forward trajectories to predict future dynamics, and the other generating backward trajectories to trace essential history transitions.BiTrajDiff can efficiently leverage critical states as anchors to expand into potentially valuable yet underexplored regions of the state space, thereby facilitating dataset diversity. Extensive experiments on the D4RL benchmark suite demonstrate that BiTrajDiff achieves superior performance compared to other advanced DA methods across various offline RL backbones.

Linfeng Wen, Chengying Gao, Changqing Zou

Content affinity loss including feature and pixel affinity is a main problem which leads to artifacts in photorealistic and video style transfer. This paper proposes a new framework named CAP-VSTNet, which consists of a new reversible residual network and an unbiased linear transform module, for versatile style transfer. This reversible residual network can not only preserve content affinity but not introduce redundant information as traditional reversible networks, and hence facilitate better stylization. Empowered by Matting Laplacian training loss which can address the pixel affinity loss problem led by the linear transform, the proposed framework is applicable and effective on versatile style transfer. Extensive experiments show that CAP-VSTNet can produce better qualitative and quantitative results in comparison with the state-of-the-art methods.

Xinyu Gao, Ziyi Yang, Yunlu Zhao et al.

A variety of Neural Radiance Fields (NeRF) methods have recently achieved remarkable success in high render speed. However, current accelerating methods are specialized and incompatible with various implicit methods, preventing real-time composition over various types of NeRF works. Because NeRF relies on sampling along rays, it is possible to provide general guidance for acceleration. To that end, we propose a general implicit pipeline for composing NeRF objects quickly. Our method enables the casting of dynamic shadows within or between objects using analytical light sources while allowing multiple NeRF objects to be seamlessly placed and rendered together with any arbitrary rigid transformations. Mainly, our work introduces a new surface representation known as Neural Depth Fields (NeDF) that quickly determines the spatial relationship between objects by allowing direct intersection computation between rays and implicit surfaces. It leverages an intersection neural network to query NeRF for acceleration instead of depending on an explicit spatial structure.Our proposed method is the first to enable both the progressive and interactive composition of NeRF objects. Additionally, it also serves as a previewing plugin for a range of existing NeRF works.

Zhimin Fan, Pengpei Hong, Jie Guo et al.

Complex visual effects such as caustics are often produced by light paths containing multiple consecutive specular vertices (dubbed specular chains), which pose a challenge to unbiased estimation in Monte Carlo rendering. In this work, we study the light transport behavior within a sub-path that is comprised of a specular chain and two non-specular separators. We show that the specular manifolds formed by all the sub-paths could be exploited to provide coherence among sub-paths. By reconstructing continuous energy distributions from historical and coherent sub-paths, seed chains can be generated in the context of importance sampling and converge to admissible chains through manifold walks. We verify that importance sampling the seed chain in the continuous space reaches the goal of importance sampling the discrete admissible specular chain. Based on these observations and theoretical analyses, a progressive pipeline, manifold path guiding, is designed and implemented to importance sample challenging paths featuring long specular chains. To our best knowledge, this is the first general framework for importance sampling discrete specular chains in regular Monte Carlo rendering. Extensive experiments demonstrate that our method outperforms state-of-the-art unbiased solutions with up to 40x variance reduction, especially in typical scenes containing long specular chains and complex visibility.

Yihong Guo, Youwei Lyu, Jiajun Tang et al.

Reasoning photo retouching has gained significant traction, requiring models to analyze image defects, give reasoning processes, and execute precise retouching enhancements. However, existing approaches often rely on non-differentiable external software, creating optimization barriers and suffering from high parameter redundancy and limited generalization. To address these challenges, we propose VeraRetouch, a lightweight and fully differentiable framework for multi-task photo retouching. We employ a 0.5B Vision-Language Model (VLM) as the central intelligence to formulate retouching plans based on instructions and scene semantics. Furthermore, we develop a fully differentiable Retouch Renderer that replaces external tools, enabling direct end-to-end pixel-level training through decoupled control latents for lighting, global color, and specific color adjustments. To overcome data scarcity, we introduce AetherRetouch-1M+, the first million-scale dataset for professional retouching, constructed via a new inverse degradation workflow. Furthermore, we propose DAPO-AE, a reinforcement learning post-training strategy that enhances autonomous aesthetic cognition. Extensive experiments demonstrate that VeraRetouch achieves state-of-the-art performance across multiple benchmarks while maintaining a significantly smaller footprint, enabling mobile deployment. Our code and models are publicly available at https://github.com/OpenVeraTeam/VeraRetouch.

Xinlei Yu, Changmiao Wang, Hui Jin et al.

Multi-organ medical segmentation is a crucial component of medical image processing, essential for doctors to make accurate diagnoses and develop effective treatment plans. Despite significant progress in this field, current multi-organ segmentation models often suffer from inaccurate details, dependence on geometric prompts and loss of spatial information. Addressing these challenges, we introduce a novel model named CRISP-SAM2 with CRoss-modal Interaction and Semantic Prompting based on SAM2. This model represents a promising approach to multi-organ medical segmentation guided by textual descriptions of organs. Our method begins by converting visual and textual inputs into cross-modal contextualized semantics using a progressive cross-attention interaction mechanism. These semantics are then injected into the image encoder to enhance the detailed understanding of visual information. To eliminate reliance on geometric prompts, we use a semantic prompting strategy, replacing the original prompt encoder to sharpen the perception of challenging targets. In addition, a similarity-sorting self-updating strategy for memory and a mask-refining process is applied to further adapt to medical imaging and enhance localized details. Comparative experiments conducted on seven public datasets indicate that CRISP-SAM2 outperforms existing models. Extensive analysis also demonstrates the effectiveness of our method, thereby confirming its superior performance, especially in addressing the limitations mentioned earlier. Our code is available at: https://github.com/YU-deep/CRISP_SAM2.git.

Jie Zhang, Minghui Nie, Changqing Zou et al.

Although supervised deep normal estimators have recently shown impressive results on synthetic benchmarks, their performance deteriorates significantly in real-world scenarios due to the domain gap between synthetic and real data. Building high-quality real training data to boost those supervised methods is not trivial because point-wise annotation of normals for varying-scale real-world 3D scenes is a tedious and expensive task. This paper introduces PointNorm-Net, the first self-supervised deep learning framework to tackle this challenge. The key novelty of PointNorm-Net is a three-stage multi-modal normal distribution estimation paradigm that can be integrated into either deep or traditional optimization-based normal estimation frameworks. Extensive experiments show that our method achieves superior generalization and outperforms state-of-the-art conventional and deep learning approaches across three real-world datasets that exhibit distinct characteristics compared to the synthetic training data.

Qiyuan Zhang, Biao Gong, Shuai Tan et al.

Physical principles are fundamental to realistic visual simulation, but remain a significant oversight in transformer-based video generation. This gap highlights a critical limitation in rendering rigid body motion, a core tenet of classical mechanics. While computer graphics and physics-based simulators can easily model such collisions using Newton formulas, modern pretrain-finetune paradigms discard the concept of object rigidity during pixel-level global denoising. Even perfectly correct mathematical constraints are treated as suboptimal solutions (i.e., conditions) during model optimization in post-training, fundamentally limiting the physical realism of generated videos. Motivated by these considerations, we introduce, for the first time, a physics-aware reinforcement learning paradigm for video generation models that enforces physical collision rules directly in high-dimensional spaces, ensuring the physics knowledge is strictly applied rather than treated as conditions. Subsequently, we extend this paradigm to a unified framework, termed Mimicry-Discovery Cycle (MDcycle), which allows substantial fine-tuning while fully preserving the model's ability to leverage physics-grounded feedback. To validate our approach, we construct new benchmark PhysRVGBench and perform extensive qualitative and quantitative experiments to thoroughly assess its effectiveness.

Linwei Dong, Ruoyu Guo, Ge Bai et al.

Diffusion distillation, exemplified by Distribution Matching Distillation (DMD), has shown great promise in few-step generation but often sacrifices quality for sampling speed. While integrating Reinforcement Learning (RL) into distillation offers potential, a naive fusion of these two objectives relies on suboptimal raw sample evaluation. This sample-based scoring creates inherent conflicts with the distillation trajectory and produces unreliable rewards due to the noisy nature of early-stage generation. To overcome these limitations, we propose GDMD, a novel framework that redefines the reward mechanism by prioritizing distillation gradients over raw pixel outputs as the primary signal for optimization. By reinterpreting the DMD gradients as implicit target tensors, our framework enables existing reward models to directly evaluate the quality of distillation updates. This gradient-level guidance functions as an adaptive weighting that synchronizes the RL policy with the distillation objective, effectively neutralizing optimization divergence. Empirical results show that GDMD sets a new SOTA for few-step generation. Specifically, our 4-step models outperform the quality of their multi-step teacher and substantially exceed previous DMDR results in GenEval and human-preference metrics, exhibiting strong scalability potential.

Enya Shen, Zhidong Cao, Changqing Zou et al.

Network embedding aims to find a way to encode network by learning an embedding vector for each node in the network. The network often has property information which is highly informative with respect to the node's position and role in the network. Most network embedding methods fail to utilize this information during network representation learning. In this paper, we propose a novel framework, FANE, to integrate structure and property information in the network embedding process. In FANE, we design a network to unify heterogeneity of the two information sources, and define a new random walking strategy to leverage property information and make the two information compensate. FANE is conceptually simple and empirically powerful. It improves over the state-of-the-art methods on Cora dataset classification task by over 5%, more than 10% on WebKB dataset classification task. Experiments also show that the results improve more than the state-of-the-art methods as increasing training size. Moreover, qualitative visualization show that our framework is helpful in network property information exploration. In all, we present a new way for efficiently learning state-of-the-art task-independent representations in complex attributed networks. The source code and datasets of this paper can be obtained from https://github.com/GraphWorld/FANE.

Changqing Zou, Qian Yu, Ruofei Du et al.

We contribute the first large-scale dataset of scene sketches, SketchyScene, with the goal of advancing research on sketch understanding at both the object and scene level. The dataset is created through a novel and carefully designed crowdsourcing pipeline, enabling users to efficiently generate large quantities of realistic and diverse scene sketches. SketchyScene contains more than 29,000 scene-level sketches, 7,000+ pairs of scene templates and photos, and 11,000+ object sketches. All objects in the scene sketches have ground-truth semantic and instance masks. The dataset is also highly scalable and extensible, easily allowing augmenting and/or changing scene composition. We demonstrate the potential impact of SketchyScene by training new computational models for semantic segmentation of scene sketches and showing how the new dataset enables several applications including image retrieval, sketch colorization, editing, and captioning, etc. The dataset and code can be found at https://github.com/SketchyScene/SketchyScene.

Linwei Dong, Qingnan Fan, Yihong Guo et al.

Pre-trained text-to-image diffusion models are increasingly applied to real-world image super-resolution (Real-ISR) task. Given the iterative refinement nature of diffusion models, most existing approaches are computationally expensive. While methods such as SinSR and OSEDiff have emerged to condense inference steps via distillation, their performance in image restoration or details recovery is not satisfied. To address this, we propose TSD-SR, a novel distillation framework specifically designed for real-world image super-resolution, aiming to construct an efficient and effective one-step model. We first introduce the Target Score Distillation, which leverages the priors of diffusion models and real image references to achieve more realistic image restoration. Secondly, we propose a Distribution-Aware Sampling Module to make detail-oriented gradients more readily accessible, addressing the challenge of recovering fine details. Extensive experiments demonstrate that our TSD-SR has superior restoration results (most of the metrics perform the best) and the fastest inference speed (e.g. 40 times faster than SeeSR) compared to the past Real-ISR approaches based on pre-trained diffusion priors.

Linwei Dong, Ruoyu Guo, Ge Bai et al.

The slow iterative sampling nature remains a major bottleneck for the practical deployment of diffusion and flow-based generative models. While consistency models (CMs) represent a state-of-the-art distillation-based approach for efficient generation, their large-scale application is still limited by two key issues: training instability and inflexible sampling. Existing methods seek to mitigate these problems through architectural adjustments or regularized objectives, yet overlook the critical reliance on trajectory selection. In this work, we first conduct an analysis on these two limitations: training instability originates from loss divergence induced by unstable self-supervised term, whereas sampling inflexibility arises from error accumulation. Based on these insights and analysis, we propose the Dual-End Consistency Model (DE-CM) that selects vital sub-trajectory clusters to achieve stable and effective training. DE-CM decomposes the PF-ODE trajectory and selects three critical sub-trajectories as optimization targets. Specifically, our approach leverages continuous-time CMs objectives to achieve few-step distillation and utilizes flow matching as a boundary regularizer to stabilize the training process. Furthermore, we propose a novel noise-to-noisy (N2N) mapping that can map noise to any point, thereby alleviating the error accumulation in the first step. Extensive experimental results show the effectiveness of our method: it achieves a state-of-the-art FID score of 1.70 in one-step generation on the ImageNet 256x256 dataset, outperforming existing CM-based one-step approaches.

Huaize Liu, Wenzhang Sun, Donglin Di et al.

The generation of talking avatars has achieved significant advancements in precise audio synchronization. However, crafting lifelike talking head videos requires capturing a broad spectrum of emotions and subtle facial expressions. Current methods face fundamental challenges: a) the absence of frameworks for modeling single basic emotional expressions, which restricts the generation of complex emotions such as compound emotions; b) the lack of comprehensive datasets rich in human emotional expressions, which limits the potential of models. To address these challenges, we propose the following innovations: 1) the Mixture of Emotion Experts (MoEE) model, which decouples six fundamental emotions to enable the precise synthesis of both singular and compound emotional states; 2) the DH-FaceEmoVid-150 dataset, specifically curated to include six prevalent human emotional expressions as well as four types of compound emotions, thereby expanding the training potential of emotion-driven models. Furthermore, to enhance the flexibility of emotion control, we propose an emotion-to-latents module that leverages multimodal inputs, aligning diverse control signals-such as audio, text, and labels-to ensure more varied control inputs as well as the ability to control emotions using audio alone. Through extensive quantitative and qualitative evaluations, we demonstrate that the MoEE framework, in conjunction with the DH-FaceEmoVid-150 dataset, excels in generating complex emotional expressions and nuanced facial details, setting a new benchmark in the field. These datasets will be publicly released.

Yibo Zhang, Lihong Wang, Changqing Zou et al.

3D sketches are widely used for visually representing the 3D shape and structure of objects or scenes. However, the creation of 3D sketch often requires users to possess professional artistic skills. Existing research efforts primarily focus on enhancing the ability of interactive sketch generation in 3D virtual systems. In this work, we propose Diff3DS, a novel differentiable rendering framework for generating view-consistent 3D sketch by optimizing 3D parametric curves under various supervisions. Specifically, we perform perspective projection to render the 3D rational Bézier curves into 2D curves, which are subsequently converted to a 2D raster image via our customized differentiable rasterizer. Our framework bridges the domains of 3D sketch and raster image, achieving end-toend optimization of 3D sketch through gradients computed in the 2D image domain. Our Diff3DS can enable a series of novel 3D sketch generation tasks, including textto-3D sketch and image-to-3D sketch, supported by the popular distillation-based supervision, such as Score Distillation Sampling (SDS). Extensive experiments have yielded promising results and demonstrated the potential of our framework. Project page is at https://yiboz2001.github.io/Diff3DS/.

Songchun Zhang, Huiyao Xu, Sitong Guo et al.

Novel view synthesis (NVS) boosts immersive experiences in computer vision and graphics. Existing techniques, though progressed, rely on dense multi-view observations, restricting their application. This work takes on the challenge of reconstructing photorealistic 3D scenes from sparse or single-view inputs. We introduce SpatialCrafter, a framework that leverages the rich knowledge in video diffusion models to generate plausible additional observations, thereby alleviating reconstruction ambiguity. Through a trainable camera encoder and an epipolar attention mechanism for explicit geometric constraints, we achieve precise camera control and 3D consistency, further reinforced by a unified scale estimation strategy to handle scale discrepancies across datasets. Furthermore, by integrating monocular depth priors with semantic features in the video latent space, our framework directly regresses 3D Gaussian primitives and efficiently processes long-sequence features using a hybrid network structure. Extensive experiments show our method enhances sparse view reconstruction and restores the realistic appearance of 3D scenes.

Miaowei Wang, Yibo Zhang, Rui Ma et al.

We present DecoupledGaussian, a novel system that decouples static objects from their contacted surfaces captured in-the-wild videos, a key prerequisite for realistic Newtonian-based physical simulations. Unlike prior methods focused on synthetic data or elastic jittering along the contact surface, which prevent objects from fully detaching or moving independently, DecoupledGaussian allows for significant positional changes without being constrained by the initial contacted surface. Recognizing the limitations of current 2D inpainting tools for restoring 3D locations, our approach proposes joint Poisson fields to repair and expand the Gaussians of both objects and contacted scenes after separation. This is complemented by a multi-carve strategy to refine the object's geometry. Our system enables realistic simulations of decoupling motions, collisions, and fractures driven by user-specified impulses, supporting complex interactions within and across multiple scenes. We validate DecoupledGaussian through a comprehensive user study and quantitative benchmarks. This system enhances digital interaction with objects and scenes in real-world environments, benefiting industries such as VR, robotics, and autonomous driving. Our project page is at: https://wangmiaowei.github.io/DecoupledGaussian.github.io/.

Linwei Dong, Qingnan Fan, Yuhang Yu et al.

Real-world image super-resolution (Real-ISR) focuses on recovering high-quality images from low-resolution inputs that suffer from complex degradations like noise, blur, and compression. Recently, diffusion models (DMs) have shown great potential in this area by leveraging strong generative priors to restore fine details. However, their iterative denoising process incurs high computational overhead, posing challenges for real-time applications. Although one-step distillation methods, such as OSEDiff and TSD-SR, offer faster inference, they remain fundamentally constrained by their large, over-parameterized model architectures. In this work, we present TinySR, a compact yet effective diffusion model specifically designed for Real-ISR that achieves real-time performance while maintaining perceptual quality. We introduce a Dynamic Inter-block Activation and an Expansion-Corrosion Strategy to facilitate more effective decision-making in depth pruning. We achieve VAE compression through channel pruning, attention removal and lightweight SepConv. We eliminate time- and prompt-related modules and perform pre-caching techniques to further speed up the model. TinySR significantly reduces computational cost and model size, achieving up to 5.68x speedup and 83% parameter reduction compared to its teacher TSD-SR, while still providing high quality results.

Hao Sun, Fenggen Yu, Huiyao Xu et al.

Novel view synthesis (NVS) in low-light scenes remains a significant challenge due to degraded inputs characterized by severe noise, low dynamic range (LDR) and unreliable initialization. While recent NeRF-based approaches have shown promising results, most suffer from high computational costs, and some rely on carefully captured or pre-processed data--such as RAW sensor inputs or multi-exposure sequences--which severely limits their practicality. In contrast, 3D Gaussian Splatting (3DGS) enables real-time rendering with competitive visual fidelity; however, existing 3DGS-based methods struggle with low-light sRGB inputs, resulting in unstable Gaussian initialization and ineffective noise suppression. To address these challenges, we propose LL-Gaussian, a novel framework for 3D reconstruction and enhancement from low-light sRGB images, enabling pseudo normal-light novel view synthesis. Our method introduces three key innovations: 1) an end-to-end Low-Light Gaussian Initialization Module (LLGIM) that leverages dense priors from learning-based MVS approach to generate high-quality initial point clouds; 2) a dual-branch Gaussian decomposition model that disentangles intrinsic scene properties (reflectance and illumination) from transient interference, enabling stable and interpretable optimization; 3) an unsupervised optimization strategy guided by both physical constrains and diffusion prior to jointly steer decomposition and enhancement. Additionally, we contribute a challenging dataset collected in extreme low-light environments and demonstrate the effectiveness of LL-Gaussian. Compared to state-of-the-art NeRF-based methods, LL-Gaussian achieves up to 2,000 times faster inference and reduces training time to just 2%, while delivering superior reconstruction and rendering quality.

Qiyuan Zhang, Chenyu Wu, Wenzhang Sun et al.

Recent advancements in portrait video generation have been noteworthy. However, existing methods rely heavily on human priors and pre-trained generative models, Motion representations based on human priors may introduce unrealistic motion, while methods relying on pre-trained generative models often suffer from inefficient inference. To address these challenges, we propose Semantic Latent Motion (SeMo), a compact and expressive motion representation. Leveraging this representation, our approach achieve both high-quality visual results and efficient inference. SeMo follows an effective three-step framework: Abstraction, Reasoning, and Generation. First, in the Abstraction step, we use a carefully designed Masked Motion Encoder, which leverages a self-supervised learning paradigm to compress the subject's motion state into a compact and abstract latent motion (1D token). Second, in the Reasoning step, we efficiently generate motion sequences based on the driving audio signal. Finally, in the Generation step, the motion dynamics serve as conditional information to guide the motion decoder in synthesizing realistic transitions from reference frame to target video. Thanks to the compact and expressive nature of Semantic Latent Motion, our method achieves efficient motion representation and high-quality video generation. User studies demonstrate that our approach surpasses state-of-the-art models with an 81% win rate in realism. Extensive experiments further highlight its strong compression capability, reconstruction quality, and generative potential.

Zeyu Ling, Xiaodong Gu, Jiangnan Tang et al.

We introduce SyncLipMAE, a self-supervised pretraining framework for talking-face video that learns synchronization-aware and transferable facial dynamics from unlabeled audio-visual streams. Our approach couples masked visual modeling with cross-modal contrastive alignment and employs three per-frame prompt tokens that explicitly encode the essential factors of a talking-face frame - identity, vocal motion (speech-synchronized facial dynamics), and ambient motion (audio-agnostic movements such as blinks and head pose). The contrastive objective uses time-aligned vocal-motion and audio tokens as positives and misaligned pairs as negatives, driving both modalities into a shared embedding space and yielding token-level audio-visual stream synchronization. After pretraining, the aligned audio tokens together with the visual prompt tokens (identity, vocal motion, ambient motion) form a unified interface for four disparate downstream settings: (i) audio-visual stream synchronization; (ii) facial emotion and head/face action recognition; (iii) visual speech recognition; and (iv) visual dubbing, for which we enable indistinguishable audio- or video-driven control within a single model. Across four task families that require distinct capabilities, SyncLipMAE achieves state-of-the-art results, underscoring the effectiveness of synchronization-aware, factorized self-supervised pretraining.

Huaize Liu, Wenzhang Sun, Qiyuan Zhang et al.

Recent breakthroughs in video autoencoders (Video AEs) have advanced video generation, but existing methods fail to efficiently model spatio-temporal redundancies in dynamics, resulting in suboptimal compression factors. This shortfall leads to excessive training costs for downstream tasks. To address this, we introduce Hi-VAE, an efficient video autoencoding framework that hierarchically encode coarse-to-fine motion representations of video dynamics and formulate the decoding process as a conditional generation task. Specifically, Hi-VAE decomposes video dynamics into two latent spaces: Global Motion, capturing overarching motion patterns, and Detailed Motion, encoding high-frequency spatial details. Using separate self-supervised motion encoders, we compress video latents into compact motion representations to reduce redundancy significantly. A conditional diffusion decoder then reconstructs videos by combining hierarchical global and detailed motions, enabling high-fidelity video reconstructions. Extensive experiments demonstrate that Hi-VAE achieves a high compression factor of 1428$\times$, almost 30$\times$ higher than baseline methods (e.g., Cosmos-VAE at 48$\times$), validating the efficiency of our approach. Meanwhile, Hi-VAE maintains high reconstruction quality at such high compression rates and performs effectively in downstream generative tasks. Moreover, Hi-VAE exhibits interpretability and scalability, providing new perspectives for future exploration in video latent representation and generation.

Cong Wang, Weizhe Yang, Haiping Wang et al.

Rapid reconstruction of 2D plasma profiles from line-integral measurements is important in nuclear fusion. This paper introduces a physics-informed model architecture called Onion, that can enhance the performance of models and be adapted to various backbone networks. The model under Onion incorporates physical information by a multiplication process and applies the physics-informed loss function according to the principle of line integration. Prediction results demonstrate that the additional input of physical information improves the deep learning model's ability, leading to a reduction in the average relative error E_1 between the reconstruction profiles and the target profiles by approximately 0.84x10^(-2) on synthetic datasets and about 0.06x10^(-2) on experimental datasets. Furthermore, the implementation of the Softplus activation function in the final two fully connected layers improves model performance. This enhancement results in a reduction in the E_1 by approximately 1.06x10^(-2) on synthetic datasets and about 0.11x10^(-2) on experimental datasets. The incorporation of the physics-informed loss function has been shown to correct the model's predictions, bringing the back-projections closer to the actual inputs and reducing the errors associated with inversion algorithms. Besides, we have developed a synthetic data model to generate customized line-integral diagnostic datasets and have also collected soft x-ray diagnostic datasets from EAST and HL-2A. This study achieves reductions in reconstruction errors, and accelerates the development of surrogate models in fusion research.

Zeyu Ling, Bo Han, Shiyang Li et al.

Large language models (LLMs) are, by design, inherently capable of multi-task learning: through a unified next-token prediction paradigm, they can naturally address a wide variety of downstream tasks. Prior work in the motion domain has demonstrated some generality by adapting LLMs via a Motion Tokenizer coupled with an autoregressive Transformer to generate and understand human motion. However, this generality remains limited in scope and yields only modest performance gains. We introduce VersatileMotion, a unified multimodal motion LLM that combines a novel motion tokenizer, integrating VQ-VAE with flow matching, and an autoregressive transformer backbone to seamlessly support at least nine distinct motion-related tasks. VersatileMotion is the first method to handle single-agent and multi-agent motions in a single framework and enable cross-modal conversion between motion, text, music, and speech, achieving state-of-the-art performance on seven of these tasks. Each sequence in MotionHub may include one or more of the following annotations: natural-language captions, music or audio clips, speech transcripts, and multi-agent interaction data. To facilitate evaluation, we define and release benchmark splits covering nine core tasks. Extensive experiments demonstrate the superior performance, versatility, and potential of VersatileMotion as a foundational model for future understanding and generation of motion.

Shuaiying Hou, Hongyu Tao, Junheng Fang et al.

Learning 3D human motion from 2D inputs is a fundamental task in the realms of computer vision and computer graphics. Many previous methods grapple with this inherently ambiguous task by introducing motion priors into the learning process. However, these approaches face difficulties in defining the complete configurations of such priors or training a robust model. In this paper, we present the Video-to-Motion Generator (VTM), which leverages motion priors through cross-modal latent feature space alignment between 3D human motion and 2D inputs, namely videos and 2D keypoints. To reduce the complexity of modeling motion priors, we model the motion data separately for the upper and lower body parts. Additionally, we align the motion data with a scale-invariant virtual skeleton to mitigate the interference of human skeleton variations to the motion priors. Evaluated on AIST++, the VTM showcases state-of-the-art performance in reconstructing 3D human motion from monocular videos. Notably, our VTM exhibits the capabilities for generalization to unseen view angles and in-the-wild videos.

Frank Zhang, Yibo Zhang, Quan Zheng et al.

Text-driven 3D scene generation techniques have made rapid progress in recent years. Their success is mainly attributed to using existing generative models to iteratively perform image warping and inpainting to generate 3D scenes. However, these methods heavily rely on the outputs of existing models, leading to error accumulation in geometry and appearance that prevent the models from being used in various scenarios (e.g., outdoor and unreal scenarios). To address this limitation, we generatively refine the newly generated local views by querying and aggregating global 3D information, and then progressively generate the 3D scene. Specifically, we employ a tri-plane features-based NeRF as a unified representation of the 3D scene to constrain global 3D consistency, and propose a generative refinement network to synthesize new contents with higher quality by exploiting the natural image prior from 2D diffusion model as well as the global 3D information of the current scene. Our extensive experiments demonstrate that, in comparison to previous methods, our approach supports wide variety of scene generation and arbitrary camera trajectories with improved visual quality and 3D consistency.

Zeyu Ruan, Changqing Zou, Longhai Wu et al.

Three-dimensional face dense alignment and reconstruction in the wild is a challenging problem as partial facial information is commonly missing in occluded and large pose face images. Large head pose variations also increase the solution space and make the modeling more difficult. Our key idea is to model occlusion and pose to decompose this challenging task into several relatively more manageable subtasks. To this end, we propose an end-to-end framework, termed as Self-aligned Dual face Regression Network (SADRNet), which predicts a pose-dependent face, a pose-independent face. They are combined by an occlusion-aware self-alignment to generate the final 3D face. Extensive experiments on two popular benchmarks, AFLW2000-3D and Florence, demonstrate that the proposed method achieves significant superior performance over existing state-of-the-art methods.

Xuancheng Zhang, Yutong Feng, Siqi Li et al.

This paper presents a view-guided solution for the task of point cloud completion. Unlike most existing methods directly inferring the missing points using shape priors, we address this task by introducing ViPC (view-guided point cloud completion) that takes the missing crucial global structure information from an extra single-view image. By leveraging a framework that sequentially performs effective cross-modality and cross-level fusions, our method achieves significantly superior results over typical existing solutions on a new large-scale dataset we collect for the view-guided point cloud completion task.

Siqi Li, Changqing Zou, Yipeng Li et al.

This paper presents an end-to-end 3D convolutional network named attention-based multi-modal fusion network (AMFNet) for the semantic scene completion (SSC) task of inferring the occupancy and semantic labels of a volumetric 3D scene from single-view RGB-D images. Compared with previous methods which use only the semantic features extracted from RGB-D images, the proposed AMFNet learns to perform effective 3D scene completion and semantic segmentation simultaneously via leveraging the experience of inferring 2D semantic segmentation from RGB-D images as well as the reliable depth cues in spatial dimension. It is achieved by employing a multi-modal fusion architecture boosted from 2D semantic segmentation and a 3D semantic completion network empowered by residual attention blocks. We validate our method on both the synthetic SUNCG-RGBD dataset and the real NYUv2 dataset and the results show that our method respectively achieves the gains of 2.5% and 2.6% on the synthetic SUNCG-RGBD dataset and the real NYUv2 dataset against the state-of-the-art method.

Chengying Gao, Qi Liu, Qi Xu et al.

We introduce the first method for automatic image generation from scene-level freehand sketches. Our model allows for controllable image generation by specifying the synthesis goal via freehand sketches. The key contribution is an attribute vector bridged Generative Adversarial Network called EdgeGAN, which supports high visual-quality object-level image content generation without using freehand sketches as training data. We have built a large-scale composite dataset called SketchyCOCO to support and evaluate the solution. We validate our approach on the tasks of both object-level and scene-level image generation on SketchyCOCO. Through quantitative, qualitative results, human evaluation and ablation studies, we demonstrate the method's capacity to generate realistic complex scene-level images from various freehand sketches.

Lifeng Huang, Chengying Gao, Yuyin Zhou et al.

In this paper, we study physical adversarial attacks on object detectors in the wild. Previous works mostly craft instance-dependent perturbations only for rigid or planar objects. To this end, we propose to learn an adversarial pattern to effectively attack all instances belonging to the same object category, referred to as Universal Physical Camouflage Attack (UPC). Concretely, UPC crafts camouflage by jointly fooling the region proposal network, as well as misleading the classifier and the regressor to output errors. In order to make UPC effective for non-rigid or non-planar objects, we introduce a set of transformations for mimicking deformable properties. We additionally impose optimization constraint to make generated patterns look natural to human observers. To fairly evaluate the effectiveness of different physical-world attacks, we present the first standardized virtual database, AttackScenes, which simulates the real 3D world in a controllable and reproducible environment. Extensive experiments suggest the superiority of our proposed UPC compared with existing physical adversarial attackers not only in virtual environments (AttackScenes), but also in real-world physical environments. Code and dataset are available at https://mesunhlf.github.io/index_physical.html.

Haoxuan You, Yifan Feng, Xibin Zhao et al.

Three-dimensional (3D) shape recognition has drawn much research attention in the field of computer vision. The advances of deep learning encourage various deep models for 3D feature representation. For point cloud and multi-view data, two popular 3D data modalities, different models are proposed with remarkable performance. However the relation between point cloud and views has been rarely investigated. In this paper, we introduce Point-View Relation Network (PVRNet), an effective network designed to well fuse the view features and the point cloud feature with a proposed relation score module. More specifically, based on the relation score module, the point-single-view fusion feature is first extracted by fusing the point cloud feature and each single view feature with point-singe-view relation, then the point-multi-view fusion feature is extracted by fusing the point cloud feature and the features of different number of views with point-multi-view relation. Finally, the point-single-view fusion feature and point-multi-view fusion feature are further combined together to achieve a unified representation for a 3D shape. Our proposed PVRNet has been evaluated on ModelNet40 dataset for 3D shape classification and retrieval. Experimental results indicate our model can achieve significant performance improvement compared with the state-of-the-art models.

Ning Liu, Yongchao Long, Changqing Zou et al.

We propose an attention-injective deformable convolutional network called ADCrowdNet for crowd understanding that can address the accuracy degradation problem of highly congested noisy scenes. ADCrowdNet contains two concatenated networks. An attention-aware network called Attention Map Generator (AMG) first detects crowd regions in images and computes the congestion degree of these regions. Based on detected crowd regions and congestion priors, a multi-scale deformable network called Density Map Estimator (DME) then generates high-quality density maps. With the attention-aware training scheme and multi-scale deformable convolutional scheme, the proposed ADCrowdNet achieves the capability of being more effective to capture the crowd features and more resistant to various noises. We have evaluated our method on four popular crowd counting datasets (ShanghaiTech, UCF_CC_50, WorldEXPO'10, and UCSD) and an extra vehicle counting dataset TRANCOS, and our approach beats existing state-of-the-art approaches on all of these datasets.

Lei Li, Changqing Zou, Youyi Zheng et al.

Freehand sketching is a dynamic process where points are sequentially sampled and grouped as strokes for sketch acquisition on electronic devices. To recognize a sketched object, most existing methods discard such important temporal ordering and grouping information from human and simply rasterize sketches into binary images for classification. In this paper, we propose a novel single-branch attentive network architecture RNN-Rasterization-CNN (Sketch-R2CNN for short) to fully leverage the dynamics in sketches for recognition. Sketch-R2CNN takes as input only a vector sketch with grouped sequences of points, and uses an RNN for stroke attention estimation in the vector space and a CNN for 2D feature extraction in the pixel space respectively. To bridge the gap between these two spaces in neural networks, we propose a neural line rasterization module to convert the vector sketch along with the attention estimated by RNN into a bitmap image, which is subsequently consumed by CNN. The neural line rasterization module is designed in a differentiable way to yield a unified pipeline for end-to-end learning. We perform experiments on existing large-scale sketch recognition benchmarks and show that by exploiting the sketch dynamics with the attention mechanism, our method is more robust and achieves better performance than the state-of-the-art methods.

Changqing Zou, Haoran Mo, Ruofei Du et al.

We introduce LUCSS, a language-based system for interactive col- orization of scene sketches, based on their semantic understanding. LUCSS is built upon deep neural networks trained via a large-scale repository of scene sketches and cartoon-style color images with text descriptions. It con- sists of three sequential modules. First, given a scene sketch, the segmenta- tion module automatically partitions an input sketch into individual object instances. Next, the captioning module generates the text description with spatial relationships based on the instance-level segmentation results. Fi- nally, the interactive colorization module allows users to edit the caption and produce colored images based on the altered caption. Our experiments show the effectiveness of our approach and the desirability of its compo- nents to alternative choices.

Chunyu Xie, Ce Li, Baochang Zhang et al.

Skeleton-based action recognition task is entangled with complex spatio-temporal variations of skeleton joints, and remains challenging for Recurrent Neural Networks (RNNs). In this work, we propose a temporal-then-spatial recalibration scheme to alleviate such complex variations, resulting in an end-to-end Memory Attention Networks (MANs) which consist of a Temporal Attention Recalibration Module (TARM) and a Spatio-Temporal Convolution Module (STCM). Specifically, the TARM is deployed in a residual learning module that employs a novel attention learning network to recalibrate the temporal attention of frames in a skeleton sequence. The STCM treats the attention calibrated skeleton joint sequences as images and leverages the Convolution Neural Networks (CNNs) to further model the spatial and temporal information of skeleton data. These two modules (TARM and STCM) seamlessly form a single network architecture that can be trained in an end-to-end fashion. MANs significantly boost the performance of skeleton-based action recognition and achieve the best results on four challenging benchmark datasets: NTU RGB+D, HDM05, SYSU-3D and UT-Kinect.