Fan Wang

Zelin Zang, Hao Luo, Kai Wang et al. · stanford

Unsupervised Contrastive learning has gained prominence in fields such as vision, and biology, leveraging predefined positive/negative samples for representation learning. Data augmentation, categorized into hand-designed and model-based methods, has been identified as a crucial component for enhancing contrastive learning. However, hand-designed methods require human expertise in domain-specific data while sometimes distorting the meaning of the data. In contrast, generative model-based approaches usually require supervised or large-scale external data, which has become a bottleneck constraining model training in many domains. To address the problems presented above, this paper proposes DiffAug, a novel unsupervised contrastive learning technique with diffusion mode-based positive data generation. DiffAug consists of a semantic encoder and a conditional diffusion model; the conditional diffusion model generates new positive samples conditioned on the semantic encoding to serve the training of unsupervised contrast learning. With the help of iterative training of the semantic encoder and diffusion model, DiffAug improves the representation ability in an uninterrupted and unsupervised manner. Experimental evaluations show that DiffAug outperforms hand-designed and SOTA model-based augmentation methods on DNA sequence, visual, and bio-feature datasets. The code for review is released at \url{https://github.com/zangzelin/code_diffaug}.

Shuting He, Weihua Chen, Kai Wang et al. · stanford

Person Re-identification (ReID) plays a more and more crucial role in recent years with a wide range of applications. Existing ReID methods are suffering from the challenges of misalignment and occlusions, which degrade the performance dramatically. Most methods tackle such challenges by utilizing external tools to locate body parts or exploiting matching strategies. Nevertheless, the inevitable domain gap between the datasets utilized for external tools and the ReID datasets and the complicated matching process make these methods unreliable and sensitive to noises. In this paper, we propose a Region Generation and Assessment Network (RGANet) to effectively and efficiently detect the human body regions and highlight the important regions. In the proposed RGANet, we first devise a Region Generation Module (RGM) which utilizes the pre-trained CLIP to locate the human body regions using semantic prototypes extracted from text descriptions. Learnable prompt is designed to eliminate domain gap between CLIP datasets and ReID datasets. Then, to measure the importance of each generated region, we introduce a Region Assessment Module (RAM) that assigns confidence scores to different regions and reduces the negative impact of the occlusion regions by lower scores. The RAM consists of a discrimination-aware indicator and an invariance-aware indicator, where the former indicates the capability to distinguish from different identities and the latter represents consistency among the images of the same class of human body regions. Extensive experimental results for six widely-used benchmarks including three tasks (occluded, partial, and holistic) demonstrate the superiority of RGANet against state-of-the-art methods.

Bo Dong, Pichao Wang, Fan Wang

Existing semantic segmentation works have been mainly focused on designing effective decoders; however, the computational load introduced by the overall structure has long been ignored, which hinders their applications on resource-constrained hardwares. In this paper, we propose a head-free lightweight architecture specifically for semantic segmentation, named Adaptive Frequency Transformer. It adopts a parallel architecture to leverage prototype representations as specific learnable local descriptions which replaces the decoder and preserves the rich image semantics on high-resolution features. Although removing the decoder compresses most of the computation, the accuracy of the parallel structure is still limited by low computational resources. Therefore, we employ heterogeneous operators (CNN and Vision Transformer) for pixel embedding and prototype representations to further save computational costs. Moreover, it is very difficult to linearize the complexity of the vision Transformer from the perspective of spatial domain. Due to the fact that semantic segmentation is very sensitive to frequency information, we construct a lightweight prototype learning block with adaptive frequency filter of complexity $O(n)$ to replace standard self attention with $O(n^{2})$. Extensive experiments on widely adopted datasets demonstrate that our model achieves superior accuracy while retaining only 3M parameters. On the ADE20K dataset, our model achieves 41.8 mIoU and 4.6 GFLOPs, which is 4.4 mIoU higher than Segformer, with 45% less GFLOPs. On the Cityscapes dataset, our model achieves 78.7 mIoU and 34.4 GFLOPs, which is 2.5 mIoU higher than Segformer with 72.5% less GFLOPs. Code is available at https://github.com/dongbo811/AFFormer.

Shuning Chang, Pichao Wang, Ming Lin et al.

The quadratic computational complexity to the number of tokens limits the practical applications of Vision Transformers (ViTs). Several works propose to prune redundant tokens to achieve efficient ViTs. However, these methods generally suffer from (i) dramatic accuracy drops, (ii) application difficulty in the local vision transformer, and (iii) non-general-purpose networks for downstream tasks. In this work, we propose a novel Semantic Token ViT (STViT), for efficient global and local vision transformers, which can also be revised to serve as backbone for downstream tasks. The semantic tokens represent cluster centers, and they are initialized by pooling image tokens in space and recovered by attention, which can adaptively represent global or local semantic information. Due to the cluster properties, a few semantic tokens can attain the same effect as vast image tokens, for both global and local vision transformers. For instance, only 16 semantic tokens on DeiT-(Tiny,Small,Base) can achieve the same accuracy with more than 100% inference speed improvement and nearly 60% FLOPs reduction; on Swin-(Tiny,Small,Base), we can employ 16 semantic tokens in each window to further speed it up by around 20% with slight accuracy increase. Besides great success in image classification, we also extend our method to video recognition. In addition, we design a STViT-R(ecover) network to restore the detailed spatial information based on the STViT, making it work for downstream tasks, which is powerless for previous token sparsification methods. Experiments demonstrate that our method can achieve competitive results compared to the original networks in object detection and instance segmentation, with over 30% FLOPs reduction for backbone. Code is available at http://github.com/changsn/STViT-R

Fangjian Lin, Jianlong Yuan, Sitong Wu et al.

Vision Transformers have shown great potential in computer vision tasks. Most recent works have focused on elaborating the spatial token mixer for performance gains. However, we observe that a well-designed general architecture can significantly improve the performance of the entire backbone, regardless of which spatial token mixer is equipped. In this paper, we propose UniNeXt, an improved general architecture for the vision backbone. To verify its effectiveness, we instantiate the spatial token mixer with various typical and modern designs, including both convolution and attention modules. Compared with the architecture in which they are first proposed, our UniNeXt architecture can steadily boost the performance of all the spatial token mixers, and narrows the performance gap among them. Surprisingly, our UniNeXt equipped with naive local window attention even outperforms the previous state-of-the-art. Interestingly, the ranking of these spatial token mixers also changes under our UniNeXt, suggesting that an excellent spatial token mixer may be stifled due to a suboptimal general architecture, which further shows the importance of the study on the general architecture of vision backbone.

Chenjie Cao, Chaohui Yu, Fan Wang et al.

Novel View Synthesis (NVS) and 3D generation have recently achieved prominent improvements. However, these works mainly focus on confined categories or synthetic 3D assets, which are discouraged from generalizing to challenging in-the-wild scenes and fail to be employed with 2D synthesis directly. Moreover, these methods heavily depended on camera poses, limiting their real-world applications. To overcome these issues, we propose MVInpainter, re-formulating the 3D editing as a multi-view 2D inpainting task. Specifically, MVInpainter partially inpaints multi-view images with the reference guidance rather than intractably generating an entirely novel view from scratch, which largely simplifies the difficulty of in-the-wild NVS and leverages unmasked clues instead of explicit pose conditions. To ensure cross-view consistency, MVInpainter is enhanced by video priors from motion components and appearance guidance from concatenated reference key&value attention. Furthermore, MVInpainter incorporates slot attention to aggregate high-level optical flow features from unmasked regions to control the camera movement with pose-free training and inference. Sufficient scene-level experiments on both object-centric and forward-facing datasets verify the effectiveness of MVInpainter, including diverse tasks, such as multi-view object removal, synthesis, insertion, and replacement. The project page is https://ewrfcas.github.io/MVInpainter/.

Zizhang Wu, Yuanzhu Gan, Tianhao Xu et al.

The transformer-based semantic segmentation approaches, which divide the image into different regions by sliding windows and model the relation inside each window, have achieved outstanding success. However, since the relation modeling between windows was not the primary emphasis of previous work, it was not fully utilized. To address this issue, we propose a Graph-Segmenter, including a Graph Transformer and a Boundary-aware Attention module, which is an effective network for simultaneously modeling the more profound relation between windows in a global view and various pixels inside each window as a local one, and for substantial low-cost boundary adjustment. Specifically, we treat every window and pixel inside the window as nodes to construct graphs for both views and devise the Graph Transformer. The introduced boundary-aware attention module optimizes the edge information of the target objects by modeling the relationship between the pixel on the object's edge. Extensive experiments on three widely used semantic segmentation datasets (Cityscapes, ADE-20k and PASCAL Context) demonstrate that our proposed network, a Graph Transformer with Boundary-aware Attention, can achieve state-of-the-art segmentation performance.

Yuqian Yuan, Ronghao Dang, Long Li et al.

The emergence of multimodal large language models (MLLMs) has driven breakthroughs in egocentric vision applications. These applications necessitate persistent, context-aware understanding of objects, as users interact with tools in dynamic and cluttered environments. However, existing embodied benchmarks primarily focus on static scene exploration, emphasizing object's appearance and spatial attributes while neglecting the assessment of dynamic changes arising from users' interactions. To address this gap, we introduce EOC-Bench, an innovative benchmark designed to systematically evaluate object-centric embodied cognition in dynamic egocentric scenarios. Specially, EOC-Bench features 3,277 meticulously annotated QA pairs categorized into three temporal categories: Past, Present, and Future, covering 11 fine-grained evaluation dimensions and 3 visual object referencing types. To ensure thorough assessment, we develop a mixed-format human-in-the-loop annotation framework with four types of questions and design a novel multi-scale temporal accuracy metric for open-ended temporal evaluation. Based on EOC-Bench, we conduct comprehensive evaluations of various proprietary, open-source, and object-level MLLMs. EOC-Bench serves as a crucial tool for advancing the embodied object cognitive capabilities of MLLMs, establishing a robust foundation for developing reliable core models for embodied systems.

Zebin Yang, Sunjian Zheng, Tong Xie et al.

Object-goal navigation (ObjNav) tasks an agent with navigating to the location of a specific object in an unseen environment. Embodied agents equipped with large language models (LLMs) and online constructed navigation maps can perform ObjNav in a zero-shot manner. However, existing agents heavily rely on giant LLMs on the cloud, e.g., GPT-4, while directly switching to small LLMs, e.g., LLaMA3.2-11b, suffer from significant success rate drops due to limited model capacity for understanding complex navigation maps, which prevents deploying ObjNav on local devices. At the same time, the long prompt introduced by the navigation map description will cause high planning latency on local devices. In this paper, we propose EfficientNav to enable on-device efficient LLM-based zero-shot ObjNav. To help the smaller LLMs better understand the environment, we propose semantics-aware memory retrieval to prune redundant information in navigation maps. To reduce planning latency, we propose discrete memory caching and attention-based memory clustering to efficiently save and re-use the KV cache. Extensive experimental results demonstrate that EfficientNav achieves 11.1% improvement in success rate on HM3D benchmark over GPT-4-based baselines, and demonstrates 6.7x real-time latency reduction and 4.7x end-to-end latency reduction over GPT-4 planner. Our code will be released soon.

Zijie Wu, Chaohui Yu, Fan Wang et al.

Recent advances in 4D content generation have attracted increasing attention, yet creating high-quality animated 3D models remains challenging due to the complexity of modeling spatio-temporal distributions and the scarcity of 4D training data. In this paper, we present AnimateAnyMesh, the first feed-forward framework that enables efficient text-driven animation of arbitrary 3D meshes. Our approach leverages a novel DyMeshVAE architecture that effectively compresses and reconstructs dynamic mesh sequences by disentangling spatial and temporal features while preserving local topological structures. To enable high-quality text-conditional generation, we employ a Rectified Flow-based training strategy in the compressed latent space. Additionally, we contribute the DyMesh Dataset, containing over 4M diverse dynamic mesh sequences with text annotations. Experimental results demonstrate that our method generates semantically accurate and temporally coherent mesh animations in a few seconds, significantly outperforming existing approaches in both quality and efficiency. Our work marks a substantial step forward in making 4D content creation more accessible and practical. All the data, code, and models will be open-released.

Yuanpeng Tu, Hao Luo, Xi Chen et al.

We introduce PlayerOne, the first egocentric realistic world simulator, facilitating immersive and unrestricted exploration within vividly dynamic environments. Given an egocentric scene image from the user, PlayerOne can accurately construct the corresponding world and generate egocentric videos that are strictly aligned with the real scene human motion of the user captured by an exocentric camera. PlayerOne is trained in a coarse-to-fine pipeline that first performs pretraining on large-scale egocentric text-video pairs for coarse-level egocentric understanding, followed by finetuning on synchronous motion-video data extracted from egocentric-exocentric video datasets with our automatic construction pipeline. Besides, considering the varying importance of different components, we design a part-disentangled motion injection scheme, enabling precise control of part-level movements. In addition, we devise a joint reconstruction framework that progressively models both the 4D scene and video frames, ensuring scene consistency in the long-form video generation. Experimental results demonstrate its great generalization ability in precise control of varying human movements and worldconsistent modeling of diverse scenarios. It marks the first endeavor into egocentric real-world simulation and can pave the way for the community to delve into fresh frontiers of world modeling and its diverse applications.

Yuang Liu, Jing Wang, Qiang Zhou et al.

Numerous self-supervised learning paradigms, such as contrastive learning and masked image modeling, have been proposed to acquire powerful and general representations from unlabeled data. However, these models are commonly pretrained within their specific framework alone, failing to consider the complementary nature of visual representations. To tackle this issue, we introduce Comprehensive Distillation with Multiple Self-supervised Teachers (DMT) for pretrained model compression, which leverages the strengths of multiple off-the-shelf self-supervised models. Our experimental results on prominent benchmark datasets exhibit that the proposed method significantly surpasses state-of-the-art competitors while retaining favorable efficiency metrics. On classification tasks, our DMT framework utilizing three different self-supervised ViT-Base teachers enhances the performance of both small/tiny models and the base model itself. For dense tasks, DMT elevates the AP/mIoU of standard SSL models on MS-COCO and ADE20K datasets by 4.0%.

Shang Liu, Chenjie Cao, Chaohui Yu et al.

Despite the remarkable developments achieved by recent 3D generation works, scaling these methods to geographic extents, such as modeling thousands of square kilometers of Earth's surface, remains an open challenge. We address this through a dual innovation in data infrastructure and model architecture. First, we introduce Aerial-Earth3D, the largest 3D aerial dataset to date, consisting of 50k curated scenes (each measuring 600m x 600m) captured across the U.S. mainland, comprising 45M multi-view Google Earth frames. Each scene provides pose-annotated multi-view images, depth maps, normals, semantic segmentation, and camera poses, with explicit quality control to ensure terrain diversity. Building on this foundation, we propose EarthCrafter, a tailored framework for large-scale 3D Earth generation via sparse-decoupled latent diffusion. Our architecture separates structural and textural generation: 1) Dual sparse 3D-VAEs compress high-resolution geometric voxels and textural 2D Gaussian Splats (2DGS) into compact latent spaces, largely alleviating the costly computation suffering from vast geographic scales while preserving critical information. 2) We propose condition-aware flow matching models trained on mixed inputs (semantics, images, or neither) to flexibly model latent geometry and texture features independently. Extensive experiments demonstrate that EarthCrafter performs substantially better in extremely large-scale generation. The framework further supports versatile applications, from semantic-guided urban layout generation to unconditional terrain synthesis, while maintaining geographic plausibility through our rich data priors from Aerial-Earth3D. Our project page is available at https://whiteinblue.github.io/earthcrafter/

Yang Xue, Zijing Liu, Xiaomin Fang et al.

Protein-protein interactions (PPIs) are essentials for many biological processes where two or more proteins physically bind together to achieve their functions. Modeling PPIs is useful for many biomedical applications, such as vaccine design, antibody therapeutics, and peptide drug discovery. Pre-training a protein model to learn effective representation is critical for PPIs. Most pre-training models for PPIs are sequence-based, which naively adopt the language models used in natural language processing to amino acid sequences. More advanced works utilize the structure-aware pre-training technique, taking advantage of the contact maps of known protein structures. However, neither sequences nor contact maps can fully characterize structures and functions of the proteins, which are closely related to the PPI problem. Inspired by this insight, we propose a multimodal protein pre-training model with three modalities: sequence, structure, and function (S2F). Notably, instead of using contact maps to learn the amino acid-level rigid structures, we encode the structure feature with the topology complex of point clouds of heavy atoms. It allows our model to learn structural information about not only the backbones but also the side chains. Moreover, our model incorporates the knowledge from the functional description of proteins extracted from literature or manual annotations. Our experiments show that the S2F learns protein embeddings that achieve good performances on a variety of PPIs tasks, including cross-species PPI, antibody-antigen affinity prediction, antibody neutralization prediction for SARS-CoV-2, and mutation-driven binding affinity change prediction.

Ming Yan, Haiyang Xu, Chenliang Li et al.

The Visual Question Answering (VQA) task utilizes both visual image and language analysis to answer a textual question with respect to an image. It has been a popular research topic with an increasing number of real-world applications in the last decade. This paper describes our recent research of AliceMind-MMU (ALIbaba's Collection of Encoder-decoders from Machine IntelligeNce lab of Damo academy - MultiMedia Understanding) that obtains similar or even slightly better results than human being does on VQA. This is achieved by systematically improving the VQA pipeline including: (1) pre-training with comprehensive visual and textual feature representation; (2) effective cross-modal interaction with learning to attend; and (3) A novel knowledge mining framework with specialized expert modules for the complex VQA task. Treating different types of visual questions with corresponding expertise needed plays an important role in boosting the performance of our VQA architecture up to the human level. An extensive set of experiments and analysis are conducted to demonstrate the effectiveness of the new research work.

Siqi Bao, Bingjin Chen, Huang He et al.

In this work, we explore the application of PLATO-2 on various dialogue systems, including open-domain conversation, knowledge grounded dialogue, and task-oriented conversation. PLATO-2 is initially designed as an open-domain chatbot, trained via two-stage curriculum learning. In the first stage, a coarse-grained response generation model is learned to fit the simplified one-to-one mapping relationship. This model is applied to the task-oriented conversation, given that the semantic mappings tend to be deterministic in task completion. In the second stage, another fine-grained generation model and an evaluation model are further learned for diverse response generation and coherence estimation, respectively. With superior capability on capturing one-to-many mapping, such models are suitable for the open-domain conversation and knowledge grounded dialogue. For the comprehensive evaluation of PLATO-2, we have participated in multiple tasks of DSTC9, including interactive evaluation of open-domain conversation (Track3-task2), static evaluation of knowledge grounded dialogue (Track3-task1), and end-to-end task-oriented conversation (Track2-task1). PLATO-2 has obtained the 1st place in all three tasks, verifying its effectiveness as a unified framework for various dialogue systems.

Shuting He, Hao Luo, Pichao Wang et al.

Extracting robust feature representation is one of the key challenges in object re-identification (ReID). Although convolution neural network (CNN)-based methods have achieved great success, they only process one local neighborhood at a time and suffer from information loss on details caused by convolution and downsampling operators (e.g. pooling and strided convolution). To overcome these limitations, we propose a pure transformer-based object ReID framework named TransReID. Specifically, we first encode an image as a sequence of patches and build a transformer-based strong baseline with a few critical improvements, which achieves competitive results on several ReID benchmarks with CNN-based methods. To further enhance the robust feature learning in the context of transformers, two novel modules are carefully designed. (i) The jigsaw patch module (JPM) is proposed to rearrange the patch embeddings via shift and patch shuffle operations which generates robust features with improved discrimination ability and more diversified coverage. (ii) The side information embeddings (SIE) is introduced to mitigate feature bias towards camera/view variations by plugging in learnable embeddings to incorporate these non-visual clues. To the best of our knowledge, this is the first work to adopt a pure transformer for ReID research. Experimental results of TransReID are superior promising, which achieve state-of-the-art performance on both person and vehicle ReID benchmarks.

Fan Wang, Jiangxin Yang, Yanlong Cao et al.

Image Super-Resolution (SR) provides a promising technique to enhance the image quality of low-resolution optical sensors, facilitating better-performing target detection and autonomous navigation in a wide range of robotics applications. It is noted that the state-of-the-art SR methods are typically trained and tested using single-channel inputs, neglecting the fact that the cost of capturing high-resolution images in different spectral domains varies significantly. In this paper, we attempt to leverage complementary information from a low-cost channel (visible/depth) to boost image quality of an expensive channel (thermal) using fewer parameters. To this end, we first present an effective method to virtually generate pixel-wise aligned visible and thermal images based on real-time 3D reconstruction of multi-modal data captured at various viewpoints. Then, we design a feature-level multispectral fusion residual network model to perform high-accuracy SR of thermal images by adaptively integrating co-occurrence features presented in multispectral images. Experimental results demonstrate that this new approach can effectively alleviate the ill-posed inverse problem of image SR by taking into account complementary information from an additional low-cost channel, significantly outperforming state-of-the-art SR approaches in terms of both accuracy and efficiency.

Haojie Liu, Ming Lu, Zhan Ma et al.

Over the past two decades, traditional block-based video coding has made remarkable progress and spawned a series of well-known standards such as MPEG-4, H.264/AVC and H.265/HEVC. On the other hand, deep neural networks (DNNs) have shown their powerful capacity for visual content understanding, feature extraction and compact representation. Some previous works have explored the learnt video coding algorithms in an end-to-end manner, which show the great potential compared with traditional methods. In this paper, we propose an end-to-end deep neural video coding framework (NVC), which uses variational autoencoders (VAEs) with joint spatial and temporal prior aggregation (PA) to exploit the correlations in intra-frame pixels, inter-frame motions and inter-frame compensation residuals, respectively. Novel features of NVC include: 1) To estimate and compensate motion over a large range of magnitudes, we propose an unsupervised multiscale motion compensation network (MS-MCN) together with a pyramid decoder in the VAE for coding motion features that generates multiscale flow fields, 2) we design a novel adaptive spatiotemporal context model for efficient entropy coding for motion information, 3) we adopt nonlocal attention modules (NLAM) at the bottlenecks of the VAEs for implicit adaptive feature extraction and activation, leveraging its high transformation capacity and unequal weighting with joint global and local information, and 4) we introduce multi-module optimization and a multi-frame training strategy to minimize the temporal error propagation among P-frames. NVC is evaluated for the low-delay causal settings and compared with H.265/HEVC, H.264/AVC and the other learnt video compression methods following the common test conditions, demonstrating consistent gains across all popular test sequences for both PSNR and MS-SSIM distortion metrics.

Siqi Bao, Huang He, Fan Wang et al.

To build a high-quality open-domain chatbot, we introduce the effective training process of PLATO-2 via curriculum learning. There are two stages involved in the learning process. In the first stage, a coarse-grained generation model is trained to learn response generation under the simplified framework of one-to-one mapping. In the second stage, a fine-grained generative model augmented with latent variables and an evaluation model are further trained to generate diverse responses and to select the best response, respectively. PLATO-2 was trained on both Chinese and English data, whose effectiveness and superiority are verified through comprehensive evaluations, achieving new state-of-the-art results.