Haosen Yang

Haosen Yang, Deng Huang, Bin Wen et al.

Masked autoencoders (MAEs) have emerged recently as art self-supervised spatiotemporal representation learners. Inheriting from the image counterparts, however, existing video MAEs still focus largely on static appearance learning whilst are limited in learning dynamic temporal information hence less effective for video downstream tasks. To resolve this drawback, in this work we present a motion-aware variant -- MotionMAE. Apart from learning to reconstruct individual masked patches of video frames, our model is designed to additionally predict the corresponding motion structure information over time. This motion information is available at the temporal difference of nearby frames. As a result, our model can extract effectively both static appearance and dynamic motion spontaneously, leading to superior spatiotemporal representation learning capability. Extensive experiments show that our MotionMAE outperforms significantly both supervised learning baseline and state-of-the-art MAE alternatives, under both domain-specific and domain-generic pretraining-then-finetuning settings. In particular, when using ViT-B as the backbone our MotionMAE surpasses the prior art model by a margin of 1.2% on Something-Something V2 and 3.2% on UCF101 in domain-specific pretraining setting. Encouragingly, it also surpasses the competing MAEs by a large margin of over 3% on the challenging video object segmentation task. The code is available at https://github.com/happy-hsy/MotionMAE.

Boyang Xia, Wenhao Wu, Haoran Wang et al. · amazon-science

It is challenging for artificial intelligence systems to achieve accurate video recognition under the scenario of low computation costs. Adaptive inference based efficient video recognition methods typically preview videos and focus on salient parts to reduce computation costs. Most existing works focus on complex networks learning with video classification based objectives. Taking all frames as positive samples, few of them pay attention to the discrimination between positive samples (salient frames) and negative samples (non-salient frames) in supervisions. To fill this gap, in this paper, we propose a novel Non-saliency Suppression Network (NSNet), which effectively suppresses the responses of non-salient frames. Specifically, on the frame level, effective pseudo labels that can distinguish between salient and non-salient frames are generated to guide the frame saliency learning. On the video level, a temporal attention module is learned under dual video-level supervisions on both the salient and the non-salient representations. Saliency measurements from both two levels are combined for exploitation of multi-granularity complementary information. Extensive experiments conducted on four well-known benchmarks verify our NSNet not only achieves the state-of-the-art accuracy-efficiency trade-off but also present a significantly faster (2.4~4.3x) practical inference speed than state-of-the-art methods. Our project page is at https://lawrencexia2008.github.io/projects/nsnet .

Xi Chen, Haosen Yang, Sheng Jin et al.

Open-vocabulary segmentation poses significant challenges, as it requires segmenting and recognizing objects across an open set of categories in unconstrained environments. Building on the success of powerful vision-language (ViL) foundation models, such as CLIP, recent efforts sought to harness their zero-short capabilities to recognize unseen categories. Despite notable performance improvements, these models still encounter the critical issue of generating precise mask proposals for unseen categories and scenarios, resulting in inferior segmentation performance eventually. To address this challenge, we introduce a novel approach, FrozenSeg, designed to integrate spatial knowledge from a localization foundation model (e.g., SAM) and semantic knowledge extracted from a ViL model (e.g., CLIP), in a synergistic framework. Taking the ViL model's visual encoder as the feature backbone, we inject the space-aware feature into the learnable queries and CLIP features within the transformer decoder. In addition, we devise a mask proposal ensemble strategy for further improving the recall rate and mask quality. To fully exploit pre-trained knowledge while minimizing training overhead, we freeze both foundation models, focusing optimization efforts solely on a lightweight transformer decoder for mask proposal generation-the performance bottleneck. Extensive experiments demonstrate that FrozenSeg advances state-of-the-art results across various segmentation benchmarks, trained exclusively on COCO panoptic data, and tested in a zero-shot manner. Code is available at https://github.com/chenxi52/FrozenSeg.

Haosen Yang, Chuofan Ma, Bin Wen et al.

Understanding the semantics of individual regions or patches of unconstrained images, such as open-world object detection, remains a critical yet challenging task in computer vision. Building on the success of powerful image-level vision-language (ViL) foundation models like CLIP, recent efforts have sought to harness their capabilities by either training a contrastive model from scratch with an extensive collection of region-label pairs or aligning the outputs of a detection model with image-level representations of region proposals. Despite notable progress, these approaches are plagued by computationally intensive training requirements, susceptibility to data noise, and deficiency in contextual information. To address these limitations, we explore the synergistic potential of off-the-shelf foundation models, leveraging their respective strengths in localization and semantics. We introduce a novel, generic, and efficient architecture, named RegionSpot, designed to integrate position-aware localization knowledge from a localization foundation model (e.g., SAM) with semantic information from a ViL model (e.g., CLIP). To fully exploit pretrained knowledge while minimizing training overhead, we keep both foundation models frozen, focusing optimization efforts solely on a lightweight attention-based knowledge integration module. Extensive experiments in open-world object recognition show that our RegionSpot achieves significant performance gain over prior alternatives, along with substantial computational savings (e.g., training our model with 3 million data in a single day using 8 V100 GPUs). RegionSpot outperforms GLIP-L by 2.9 in mAP on LVIS val set, with an even larger margin of 13.1 AP for more challenging and rare categories, and a 2.5 AP increase on ODinW. Furthermore, it exceeds GroundingDINO-L by 11.0 AP for rare categories on the LVIS minival set.

Swapnil Bhosale, Haosen Yang, Diptesh Kanojia et al.

Audio-Visual Segmentation (AVS) aims to precisely outline audible objects in a visual scene at the pixel level. Existing AVS methods require fine-grained annotations of audio-mask pairs in supervised learning fashion. This limits their scalability since it is time consuming and tedious to acquire such cross-modality pixel level labels. To overcome this obstacle, in this work we introduce unsupervised audio-visual segmentation with no need for task-specific data annotations and model training. For tackling this newly proposed problem, we formulate a novel Cross-Modality Semantic Filtering (CMSF) approach to accurately associate the underlying audio-mask pairs by leveraging the off-the-shelf multi-modal foundation models (e.g., detection [1], open-world segmentation [2] and multi-modal alignment [3]). Guiding the proposal generation by either audio or visual cues, we design two training-free variants: AT-GDINO-SAM and OWOD-BIND. Extensive experiments on the AVS-Bench dataset show that our unsupervised approach can perform well in comparison to prior art supervised counterparts across complex scenarios with multiple auditory objects. Particularly, in situations where existing supervised AVS methods struggle with overlapping foreground objects, our models still excel in accurately segmenting overlapped auditory objects. Our code will be publicly released.

Xi Chen, Haosen Yang, Huicong Zhang et al.

Source-free unsupervised domain adaptation (SFUDA) aims to enable the utilization of a pre-trained source model in an unlabeled target domain without access to source data. Self-training is a way to solve SFUDA, where confident target samples are iteratively selected as pseudo-labeled samples to guide target model learning. However, prior heuristic noisy pseudo-label filtering methods all involve introducing extra models, which are sensitive to model assumptions and may introduce additional errors or mislabeling. In this work, we propose a method called Uncertainty-aware Pseudo-label-filtering Adaptation (UPA) to efficiently address this issue in a coarse-to-fine manner. Specially, we first introduce a sample selection module named Adaptive Pseudo-label Selection (APS), which is responsible for filtering noisy pseudo labels. The APS utilizes a simple sample uncertainty estimation method by aggregating knowledge from neighboring samples and confident samples are selected as clean pseudo-labeled. Additionally, we incorporate Class-Aware Contrastive Learning (CACL) to mitigate the memorization of pseudo-label noise by learning robust pair-wise representation supervised by pseudo labels. Through extensive experiments conducted on three widely used benchmarks, we demonstrate that our proposed method achieves competitive performance on par with state-of-the-art SFUDA methods. Code is available at https://github.com/chenxi52/UPA.

Wenqing Wang, Haosen Yang, Josef Kittler et al.

The creation of 3D human face avatars from a single unconstrained image is a fundamental task that underlies numerous real-world vision and graphics applications. Despite the significant progress made in generative models, existing methods are either less suited in design for human faces or fail to generalise from the restrictive training domain to unconstrained facial images. To address these limitations, we propose a novel model, Gen3D-Face, which generates 3D human faces with unconstrained single image input within a multi-view consistent diffusion framework. Given a specific input image, our model first produces multi-view images, followed by neural surface construction. To incorporate face geometry information in a generalisable manner, we utilise input-conditioned mesh estimation instead of ground-truth mesh along with synthetic multi-view training data. Importantly, we introduce a multi-view joint generation scheme to enhance appearance consistency among different views. To the best of our knowledge, this is the first attempt and benchmark for creating photorealistic 3D human face avatars from single images for generic human subject across domains. Extensive experiments demonstrate the superiority of our method over previous alternatives for out-of-domain singe image 3D face generation and top competition for in-domain setting.

Wenqing Wang, Haosen Yang, Josef Kittler et al.

Reconstructing dynamic humans interacting with real-world environments from monocular videos is an important and challenging task. Despite considerable progress in 4D neural rendering, existing approaches either model dynamic scenes holistically or model scenes and backgrounds separately aim to introduce parametric human priors. However, these approaches either neglect distinct motion characteristics of various components in scene especially human, leading to incomplete reconstructions, or ignore the information exchange between the separately modeled components, resulting in spatial inconsistencies and visual artifacts at human-scene boundaries. To address this, we propose {\bf Separate-then-Map} (StM) strategy that introduces a dedicated information mapping mechanism to bridge separately defined and optimized models. Our method employs a shared transformation function for each Gaussian attribute to unify separately modeled components, enhancing computational efficiency by avoiding exhaustive pairwise interactions while ensuring spatial and visual coherence between humans and their surroundings. Extensive experiments on monocular video datasets demonstrate that StM significantly outperforms existing state-of-the-art methods in both visual quality and rendering accuracy, particularly at challenging human-scene interaction boundaries.

Hong Liu, Haosen Yang, Federica Eduati et al.

Leveraging multimodal data, particularly the integration of whole-slide histology images (WSIs) and transcriptomic profiles, holds great promise for improving cancer survival prediction. However, excessive redundancy in multimodal data can degrade model performance. In this paper, we propose Adaptive Prototype Learning (APL), a novel and effective approach for multimodal cancer survival analysis. APL adaptively learns representative prototypes in a data-driven manner, reducing redundancy while preserving critical information. Our method employs two sets of learnable query vectors that serve as a bridge between high-dimensional representations and survival prediction, capturing task-relevant features. Additionally, we introduce a multimodal mixed self-attention mechanism to enable cross-modal interactions, further enhancing information fusion. Extensive experiments on five benchmark cancer datasets demonstrate the superiority of our approach over existing methods. The code is available at https://github.com/HongLiuuuuu/APL.

Hong Liu, Haosen Yang, Evi M. C. Huijben et al.

Tumor segmentation plays a critical role in histopathology, but it requires costly, fine-grained image-mask pairs annotated by pathologists. Thus, synthesizing histopathology data to expand the dataset is highly desirable. Previous works suffer from inaccuracies and limited diversity in image-mask pairs, both of which affect training segmentation, particularly in small-scale datasets and the inherently complex nature of histopathology images. To address this challenge, we propose PathoPainter, which reformulates image-mask pair generation as a tumor inpainting task. Specifically, our approach preserves the background while inpainting the tumor region, ensuring precise alignment between the generated image and its corresponding mask. To enhance dataset diversity while maintaining biological plausibility, we incorporate a sampling mechanism that conditions tumor inpainting on regional embeddings from a different image. Additionally, we introduce a filtering strategy to exclude uncertain synthetic regions, further improving the quality of the generated data. Our comprehensive evaluation spans multiple datasets featuring diverse tumor types and various training data scales. As a result, segmentation improved significantly with our synthetic data, surpassing existing segmentation data synthesis approaches, e.g., 75.69% -> 77.69% on CAMELYON16. The code is available at https://github.com/HongLiuuuuu/PathoPainter.

Hong Liu, Haosen Yang, Paul J. van Diest et al.

The Segment Anything Model (SAM) marks a significant advancement in segmentation models, offering robust zero-shot abilities and dynamic prompting. However, existing medical SAMs are not suitable for the multi-scale nature of whole-slide images (WSIs), restricting their effectiveness. To resolve this drawback, we present WSI-SAM, enhancing SAM with precise object segmentation capabilities for histopathology images using multi-resolution patches, while preserving its efficient, prompt-driven design, and zero-shot abilities. To fully exploit pretrained knowledge while minimizing training overhead, we keep SAM frozen, introducing only minimal extra parameters and computational overhead. In particular, we introduce High-Resolution (HR) token, Low-Resolution (LR) token and dual mask decoder. This decoder integrates the original SAM mask decoder with a lightweight fusion module that integrates features at multiple scales. Instead of predicting a mask independently, we integrate HR and LR token at intermediate layer to jointly learn features of the same object across multiple resolutions. Experiments show that our WSI-SAM outperforms state-of-the-art SAM and its variants. In particular, our model outperforms SAM by 4.1 and 2.5 percent points on a ductal carcinoma in situ (DCIS) segmentation tasks and breast cancer metastasis segmentation task (CAMELYON16 dataset). The code will be available at https://github.com/HongLiuuuuu/WSI-SAM.

Swapnil Bhosale, Haosen Yang, Diptesh Kanojia et al.

Audio-Visual Segmentation (AVS) aims to identify, at the pixel level, the object in a visual scene that produces a given sound. Current AVS methods rely on costly fine-grained annotations of mask-audio pairs, making them impractical for scalability. To address this, we introduce unsupervised AVS, eliminating the need for such expensive annotation. To tackle this more challenging problem, we propose an unsupervised learning method, named Modality Correspondence Alignment (MoCA), which seamlessly integrates off-the-shelf foundation models like DINO, SAM, and ImageBind. This approach leverages their knowledge complementarity and optimizes their joint usage for multi-modality association. Initially, we estimate positive and negative image pairs in the feature space. For pixel-level association, we introduce an audio-visual adapter and a novel pixel matching aggregation strategy within the image-level contrastive learning framework. This allows for a flexible connection between object appearance and audio signal at the pixel level, with tolerance to imaging variations such as translation and rotation. Extensive experiments on the AVSBench (single and multi-object splits) and AVSS datasets demonstrate that our MoCA outperforms strongly designed baseline methods and approaches supervised counterparts, particularly in complex scenarios with multiple auditory objects. Notably when comparing mIoU, MoCA achieves a substantial improvement over baselines in both the AVSBench (S4: +17.24%; MS3: +67.64%) and AVSS (+19.23%) audio-visual segmentation challenges.

Haosen Yang, Adrian Bulat, Isma Hadji et al.

Diffusion models are proficient at generating high-quality images. They are however effective only when operating at the resolution used during training. Inference at a scaled resolution leads to repetitive patterns and structural distortions. Retraining at higher resolutions quickly becomes prohibitive. Thus, methods enabling pre-existing diffusion models to operate at flexible test-time resolutions are highly desirable. Previous works suffer from frequent artifacts and often introduce large latency overheads. We propose two simple modules that combine to solve these issues. We introduce a Frequency Modulation (FM) module that leverages the Fourier domain to improve the global structure consistency, and an Attention Modulation (AM) module which improves the consistency of local texture patterns, a problem largely ignored in prior works. Our method, coined Fam diffusion, can seamlessly integrate into any latent diffusion model and requires no additional training. Extensive qualitative results highlight the effectiveness of our method in addressing structural and local artifacts, while quantitative results show state-of-the-art performance. Also, our method avoids redundant inference tricks for improved consistency such as patch-based or progressive generation, leading to negligible latency overheads.

Swapnil Bhosale, Haosen Yang, Diptesh Kanojia et al.

Novel view acoustic synthesis (NVAS) aims to render binaural audio at any target viewpoint, given a mono audio emitted by a sound source at a 3D scene. Existing methods have proposed NeRF-based implicit models to exploit visual cues as a condition for synthesizing binaural audio. However, in addition to low efficiency originating from heavy NeRF rendering, these methods all have a limited ability of characterizing the entire scene environment such as room geometry, material properties, and the spatial relation between the listener and sound source. To address these issues, we propose a novel Audio-Visual Gaussian Splatting (AV-GS) model. To obtain a material-aware and geometry-aware condition for audio synthesis, we learn an explicit point-based scene representation with an audio-guidance parameter on locally initialized Gaussian points, taking into account the space relation from the listener and sound source. To make the visual scene model audio adaptive, we propose a point densification and pruning strategy to optimally distribute the Gaussian points, with the per-point contribution in sound propagation (e.g., more points needed for texture-less wall surfaces as they affect sound path diversion). Extensive experiments validate the superiority of our AV-GS over existing alternatives on the real-world RWAS and simulation-based SoundSpaces datasets.

Haosen Yang, Chenhao Zhang, Wenqing Wang et al.

Point management is critical for optimizing 3D Gaussian Splatting models, as point initiation (e.g., via structure from motion) is often distributionally inappropriate. Typically, Adaptive Density Control (ADC) algorithm is adopted, leveraging view-averaged gradient magnitude thresholding for point densification, opacity thresholding for pruning, and regular all-points opacity reset. We reveal that this strategy is limited in tackling intricate/special image regions (e.g., transparent) due to inability of identifying all 3D zones requiring point densification, and lacking an appropriate mechanism to handle ill-conditioned points with negative impacts (e.g., occlusion due to false high opacity). To address these limitations, we propose a Localized Point Management (LPM) strategy, capable of identifying those error-contributing zones in greatest need for both point addition and geometry calibration. Zone identification is achieved by leveraging the underlying multiview geometry constraints, subject to image rendering errors. We apply point densification in the identified zones and then reset the opacity of the points in front of these regions, creating a new opportunity to correct poorly conditioned points. Serving as a versatile plugin, LPM can be seamlessly integrated into existing static 3D and dynamic 4D Gaussian Splatting models with minimal additional cost. Experimental evaluations validate the efficacy of our LPM in boosting a variety of existing 3D/4D models both quantitatively and qualitatively. Notably, LPM improves both static 3DGS and dynamic SpaceTimeGS to achieve state-of-the-art rendering quality while retaining real-time speeds, excelling on challenging datasets such as Tanks & Temples and the Neural 3D Video dataset.

Haosen Yang, Wenhao Wu, Lining Wang et al.

Temporal action proposal generation (TAPG) is a challenging task that aims to locate action instances in untrimmed videos with temporal boundaries. To evaluate the confidence of proposals, the existing works typically predict action score of proposals that are supervised by the temporal Intersection-over-Union (tIoU) between proposal and the ground-truth. In this paper, we innovatively propose a general auxiliary Background Constraint idea to further suppress low-quality proposals, by utilizing the background prediction score to restrict the confidence of proposals. In this way, the Background Constraint concept can be easily plug-and-played into existing TAPG methods (e.g., BMN, GTAD). From this perspective, we propose the Background Constraint Network (BCNet) to further take advantage of the rich information of action and background. Specifically, we introduce an Action-Background Interaction module for reliable confidence evaluation, which models the inconsistency between action and background by attention mechanisms at the frame and clip levels. Extensive experiments are conducted on two popular benchmarks, i.e., ActivityNet-1.3 and THUMOS14. The results demonstrate that our method outperforms state-of-the-art methods. Equipped with the existing action classifier, our method also achieves remarkable performance on the temporal action localization task.

Lining Wang, Haosen Yang, Wenhao Wu et al.

Transformer networks are effective at modeling long-range contextual information and have recently demonstrated exemplary performance in the natural language processing domain. Conventionally, the temporal action proposal generation (TAPG) task is divided into two main sub-tasks: boundary prediction and proposal confidence prediction, which rely on the frame-level dependencies and proposal-level relationships separately. To capture the dependencies at different levels of granularity, this paper intuitively presents a unified temporal action proposal generation framework with original Transformers, called TAPG Transformer, which consists of a Boundary Transformer and a Proposal Transformer. Specifically, the Boundary Transformer captures long-term temporal dependencies to predict precise boundary information and the Proposal Transformer learns the rich inter-proposal relationships for reliable confidence evaluation. Extensive experiments are conducted on two popular benchmarks: ActivityNet-1.3 and THUMOS14, and the results demonstrate that TAPG Transformer outperforms state-of-the-art methods. Equipped with the existing action classifier, our method achieves remarkable performance on the temporal action localization task. Codes and models will be available.