Jianbo Jiao

Wenxuan Wang, Jing Wang, Chen Chen et al.

The research community has witnessed the powerful potential of self-supervised Masked Image Modeling (MIM), which enables the models capable of learning visual representation from unlabeled data. In this paper, to incorporate both the crucial global structural information and local details for dense prediction tasks, we alter the perspective to the frequency domain and present a new MIM-based framework named FreMIM for self-supervised pre-training to better accomplish medical image segmentation tasks. Based on the observations that the detailed structural information mainly lies in the high-frequency components and the high-level semantics are abundant in the low-frequency counterparts, we further incorporate multi-stage supervision to guide the representation learning during the pre-training phase. Extensive experiments on three benchmark datasets show the superior advantage of our FreMIM over previous state-of-the-art MIM methods. Compared with various baselines trained from scratch, our FreMIM could consistently bring considerable improvements to model performance. The code will be publicly available at https://github.com/Rubics-Xuan/FreMIM.

Zekang Zhang, Guangyu Gao, Youyun Tang et al.

LLM-conditioned segmentation has recently advanced rapidly by coupling large language models with iterative mask generation frameworks. However, we identify a persistent failure mode in current propose-then-select pipelines. Although high-quality mask candidates are often generated, the final prediction may fail to match the given linguistic condition. This failure arises because language semantics are typically used as static prompts or post-hoc matching signals, rather than participating in the iterative mask generation process. Through systematic analysis, we show that many errors stem from semantic misalignment rather than poor mask quality. To address this issue, we propose FlowSeg, which introduces dynamic semantic guidance via a bidirectional semantic flow between intermediate decoding states and LLM-derived condition embeddings throughout the generation process. Language conditions actively guide mask refinement at each stage, while condition embeddings are progressively updated by emerging visual evidence. This design yields semantically grounded mask representations and visually aligned language conditions, enabling more reliable matching. We further incorporate a lightweight boundary-aware refinement to selectively enhance uncertain regions without perturbing confident interiors. Extensive experiments on referring expression segmentation and reasoning segmentation tasks demonstrate that FlowSeg consistently improves language-mask alignment and achieves state-of-the-art performance. Project page: https://zkzhang98.github.io/FlowSeg_page

Wentian Xu, Jianbo Jiao

Implicit Neural Representation (INR) has been emerging in computer vision in recent years. It has been shown to be effective in parameterising continuous signals such as dense 3D models from discrete image data, e.g. the neural radius field (NeRF). However, INR is under-explored in 2D image processing tasks. Considering the basic definition and the structure of INR, we are interested in its effectiveness in low-level vision problems such as image restoration. In this work, we revisit INR and investigate its application in low-level image restoration tasks including image denoising, super-resolution, inpainting, and deblurring. Extensive experimental evaluations suggest the superior performance of INR in several low-level vision tasks with limited resources, outperforming its counterparts by over 2dB. Code and models are available at https://github.com/WenTXuL/LINR

Yanda Li, Zilong Huang, Gang Yu et al. · tencent-ai

Image matting requires high-quality pixel-level human annotations to support the training of a deep model in recent literature. Whereas such annotation is costly and hard to scale, significantly holding back the development of the research. In this work, we make the first attempt towards addressing this problem, by proposing a self-supervised pre-training approach that can leverage infinite numbers of data to boost the matting performance. The pre-training task is designed in a similar manner as image matting, where random trimap and alpha matte are generated to achieve an image disentanglement objective. The pre-trained model is then used as an initialisation of the downstream matting task for fine-tuning. Extensive experimental evaluations show that the proposed approach outperforms both the state-of-the-art matting methods and other alternative self-supervised initialisation approaches by a large margin. We also show the robustness of the proposed approach over different backbone architectures. Our project page is available at https://crystraldo.github.io/dpt_mat/.

Zekang Zhang, Guangyu Gao, Zhiyuan Fang et al.

Incremental or continual learning has been extensively studied for image classification tasks to alleviate catastrophic forgetting, a phenomenon that earlier learned knowledge is forgotten when learning new concepts. For class incremental semantic segmentation, such a phenomenon often becomes much worse due to the background shift, i.e., some concepts learned at previous stages are assigned to the background class at the current training stage, therefore, significantly reducing the performance of these old concepts. To address this issue, we propose a simple yet effective method in this paper, named Mining unseen Classes via Regional Objectness for Segmentation (MicroSeg). Our MicroSeg is based on the assumption that background regions with strong objectness possibly belong to those concepts in the historical or future stages. Therefore, to avoid forgetting old knowledge at the current training stage, our MicroSeg first splits the given image into hundreds of segment proposals with a proposal generator. Those segment proposals with strong objectness from the background are then clustered and assigned newly-defined labels during the optimization. In this way, the distribution characterizes of old concepts in the feature space could be better perceived, relieving the catastrophic forgetting caused by the background shift accordingly. Extensive experiments on Pascal VOC and ADE20K datasets show competitive results with state-of-the-art, well validating the effectiveness of the proposed MicroSeg.

Zeyu Fu, Jianbo Jiao, Robail Yasrab et al.

Self-supervised contrastive representation learning offers the advantage of learning meaningful visual representations from unlabeled medical datasets for transfer learning. However, applying current contrastive learning approaches to medical data without considering its domain-specific anatomical characteristics may lead to visual representations that are inconsistent in appearance and semantics. In this paper, we propose to improve visual representations of medical images via anatomy-aware contrastive learning (AWCL), which incorporates anatomy information to augment the positive/negative pair sampling in a contrastive learning manner. The proposed approach is demonstrated for automated fetal ultrasound imaging tasks, enabling the positive pairs from the same or different ultrasound scans that are anatomically similar to be pulled together and thus improving the representation learning. We empirically investigate the effect of inclusion of anatomy information with coarse- and fine-grained granularity, for contrastive learning and find that learning with fine-grained anatomy information which preserves intra-class difference is more effective than its counterpart. We also analyze the impact of anatomy ratio on our AWCL framework and find that using more distinct but anatomically similar samples to compose positive pairs results in better quality representations. Experiments on a large-scale fetal ultrasound dataset demonstrate that our approach is effective for learning representations that transfer well to three clinical downstream tasks, and achieves superior performance compared to ImageNet supervised and the current state-of-the-art contrastive learning methods. In particular, AWCL outperforms ImageNet supervised method by 13.8% and state-of-the-art contrastive-based method by 7.1% on a cross-domain segmentation task.

Zekang Zhang, Guangyu Gao, Jianbo Jiao et al.

Class incremental semantic segmentation aims to strike a balance between the model's stability and plasticity by maintaining old knowledge while adapting to new concepts. However, most state-of-the-art methods use the freeze strategy for stability, which compromises the model's plasticity.In contrast, releasing parameter training for plasticity could lead to the best performance for all categories, but this requires discriminative feature representation.Therefore, we prioritize the model's plasticity and propose the Contrast inter- and intra-class representations for Incremental Segmentation (CoinSeg), which pursues discriminative representations for flexible parameter tuning. Inspired by the Gaussian mixture model that samples from a mixture of Gaussian distributions, CoinSeg emphasizes intra-class diversity with multiple contrastive representation centroids. Specifically, we use mask proposals to identify regions with strong objectness that are likely to be diverse instances/centroids of a category. These mask proposals are then used for contrastive representations to reinforce intra-class diversity. Meanwhile, to avoid bias from intra-class diversity, we also apply category-level pseudo-labels to enhance category-level consistency and inter-category diversity. Additionally, CoinSeg ensures the model's stability and alleviates forgetting through a specific flexible tuning strategy. We validate CoinSeg on Pascal VOC 2012 and ADE20K datasets with multiple incremental scenarios and achieve superior results compared to previous state-of-the-art methods, especially in more challenging and realistic long-term scenarios. Code is available at https://github.com/zkzhang98/CoinSeg.

Rui Peng, Shihe Shen, Kaiqiang Xiong et al.

Reconstructing the high-fidelity surface from multi-view images, especially sparse images, is a critical and practical task that has attracted widespread attention in recent years. However, existing methods are impeded by the memory constraint or the requirement of ground-truth depths and cannot recover satisfactory geometric details. To this end, we propose SuRF, a new Surface-centric framework that incorporates a new Region sparsification based on a matching Field, achieving good trade-offs between performance, efficiency and scalability. To our knowledge, this is the first unsupervised method achieving end-to-end sparsification powered by the introduced matching field, which leverages the weight distribution to efficiently locate the boundary regions containing surface. Instead of predicting an SDF value for each voxel, we present a new region sparsification approach to sparse the volume by judging whether the voxel is inside the surface region. In this way, our model can exploit higher frequency features around the surface with less memory and computational consumption. Extensive experiments on multiple benchmarks containing complex large-scale scenes show that our reconstructions exhibit high-quality details and achieve new state-of-the-art performance, i.e., 46% improvements with 80% less memory consumption. Code is available at https://github.com/prstrive/SuRF.

Hao Ai, Wenjie Chang, Jianbo Jiao et al.

Articulated objects are ubiquitous in daily life. Our goal is to achieve a high-quality reconstruction, segmentation of independent moving parts, and analysis of articulation. Recent methods analyse two different articulation states and perform per-point part segmentation, optimising per-part articulation using cross-state correspondences, given a priori knowledge of the number of parts. Such assumptions greatly limit their applications and performance. Their robustness is reduced when objects cannot be clearly visible in both states. To address these issues, in this paper, we present a new framework, Articulation in Motion (AiM). We infer part-level decomposition, articulation kinematics, and reconstruct an interactive 3D digital replica from a user-object interaction video and a start-state scan. We propose a dual-Gaussian scene representation that is learned from an initial 3DGS scan of the object and a video that shows the movement of separate parts. It uses motion cues to segment the object into parts and assign articulation joints. Subsequently, a robust, sequential RANSAC is employed to achieve part mobility analysis without any part-level structural priors, which clusters moving primitives into rigid parts and estimates kinematics while automatically determining the number of parts. The proposed approach separates the object into parts, each represented as a 3D Gaussian set, enabling high-quality rendering. Our approach yields higher quality part segmentation than previous methods, without prior knowledge. Extensive experimental analysis on both simple and complex objects validates the effectiveness and strong generalisation ability of our approach. Project page: https://haoai-1997.github.io/AiM/.

Jiangliu Wang, Jianbo Jiao, Yibing Song et al.

This work aims to improve unsupervised audio-visual pre-training. Inspired by the efficacy of data augmentation in visual contrastive learning, we propose a novel speed co-augmentation method that randomly changes the playback speeds of both audio and video data. Despite its simplicity, the speed co-augmentation method possesses two compelling attributes: (1) it increases the diversity of audio-visual pairs and doubles the size of negative pairs, resulting in a significant enhancement in the learned representations, and (2) it changes the strict correlation between audio-visual pairs but introduces a partial relationship between the augmented pairs, which is modeled by our proposed SoftInfoNCE loss to further boost the performance. Experimental results show that the proposed method significantly improves the learned representations when compared to vanilla audio-visual contrastive learning.

Anqi Zhang, Xiaokang Ji, Guangyu Gao et al.

Recent segmentation methods leveraging Multi-modal Large Language Models (MLLMs) have shown reliable object-level segmentation and enhanced spatial perception. However, almost all previous methods predominantly rely on specialist mask decoders to interpret masks from generated segmentation-related embeddings and visual features, or incorporate multiple additional tokens to assist. This paper aims to investigate whether and how we can unlock segmentation from MLLM itSELF with 1 segmentation Embedding (SELF1E) while achieving competitive results, which eliminates the need for external decoders. To this end, our approach targets the fundamental limitation of resolution reduction in pixel-shuffled image features from MLLMs. First, we retain image features at their original uncompressed resolution, and refill them with residual features extracted from MLLM-processed compressed features, thereby improving feature precision. Subsequently, we integrate pixel-unshuffle operations on image features with and without LLM processing, respectively, to unleash the details of compressed features and amplify the residual features under uncompressed resolution, which further enhances the resolution of refilled features. Moreover, we redesign the attention mask with dual perception pathways, i.e., image-to-image and image-to-segmentation, enabling rich feature interaction between pixels and the segmentation token. Comprehensive experiments across multiple segmentation tasks validate that SELF1E achieves performance competitive with specialist mask decoder-based methods, demonstrating the feasibility of decoder-free segmentation in MLLMs. Project page: https://github.com/ANDYZAQ/SELF1E.

Wangze Xu, Huachen Gao, Shihe Shen et al.

Recently, the Neural Radiance Field (NeRF) advancement has facilitated few-shot Novel View Synthesis (NVS), which is a significant challenge in 3D vision applications. Despite numerous attempts to reduce the dense input requirement in NeRF, it still suffers from time-consumed training and rendering processes. More recently, 3D Gaussian Splatting (3DGS) achieves real-time high-quality rendering with an explicit point-based representation. However, similar to NeRF, it tends to overfit the train views for lack of constraints. In this paper, we propose \textbf{MVPGS}, a few-shot NVS method that excavates the multi-view priors based on 3D Gaussian Splatting. We leverage the recent learning-based Multi-view Stereo (MVS) to enhance the quality of geometric initialization for 3DGS. To mitigate overfitting, we propose a forward-warping method for additional appearance constraints conforming to scenes based on the computed geometry. Furthermore, we introduce a view-consistent geometry constraint for Gaussian parameters to facilitate proper optimization convergence and utilize a monocular depth regularization as compensation. Experiments show that the proposed method achieves state-of-the-art performance with real-time rendering speed. Project page: https://zezeaaa.github.io/projects/MVPGS/

Jianbo Jiao, Mohammad Alsharid, Lior Drukker et al.

Auditory and visual signals usually present together and correlate with each other, not only in natural environments but also in clinical settings. However, the audio-visual modelling in the latter case can be more challenging, due to the different sources of audio/video signals and the noise (both signal-level and semantic-level) in auditory signals -- usually speech. In this paper, we consider audio-visual modelling in a clinical setting, providing a solution to learn medical representations that benefit various clinical tasks, without human expert annotation. A simple yet effective multi-modal self-supervised learning framework is proposed for this purpose. The proposed approach is able to localise anatomical regions of interest during ultrasound imaging, with only speech audio as a reference. Experimental evaluations on a large-scale clinical multi-modal ultrasound video dataset show that the proposed self-supervised method learns good transferable anatomical representations that boost the performance of automated downstream clinical tasks, even outperforming fully-supervised solutions.

Hao Chen, Chenyuan Qu, Yu Zhang et al.

With its significant performance improvements, the deep learning paradigm has become a standard tool for modern image denoisers. While promising performance has been shown on seen noise distributions, existing approaches often suffer from generalisation to unseen noise types or general and real noise. It is understandable as the model is designed to learn paired mapping (e.g. from a noisy image to its clean version). In this paper, we instead propose to learn to disentangle the noisy image, under the intuitive assumption that different corrupted versions of the same clean image share a common latent space. A self-supervised learning framework is proposed to achieve the goal, without looking at the latent clean image. By taking two different corrupted versions of the same image as input, the proposed Multi-view Self-supervised Disentanglement (MeD) approach learns to disentangle the latent clean features from the corruptions and recover the clean image consequently. Extensive experimental analysis on both synthetic and real noise shows the superiority of the proposed method over prior self-supervised approaches, especially on unseen novel noise types. On real noise, the proposed method even outperforms its supervised counterparts by over 3 dB.

Rui Peng, Wangze Xu, Luyang Tang et al.

Despite the substantial progress of novel view synthesis, existing methods, either based on the Neural Radiance Fields (NeRF) or more recently 3D Gaussian Splatting (3DGS), suffer significant degradation when the input becomes sparse. Numerous efforts have been introduced to alleviate this problem, but they still struggle to synthesize satisfactory results efficiently, especially in the large scene. In this paper, we propose SCGaussian, a Structure Consistent Gaussian Splatting method using matching priors to learn 3D consistent scene structure. Considering the high interdependence of Gaussian attributes, we optimize the scene structure in two folds: rendering geometry and, more importantly, the position of Gaussian primitives, which is hard to be directly constrained in the vanilla 3DGS due to the non-structure property. To achieve this, we present a hybrid Gaussian representation. Besides the ordinary non-structure Gaussian primitives, our model also consists of ray-based Gaussian primitives that are bound to matching rays and whose optimization of their positions is restricted along the ray. Thus, we can utilize the matching correspondence to directly enforce the position of these Gaussian primitives to converge to the surface points where rays intersect. Extensive experiments on forward-facing, surrounding, and complex large scenes show the effectiveness of our approach with state-of-the-art performance and high efficiency. Code is available at https://github.com/prstrive/SCGaussian.

Shihe Shen, Huachen Gao, Wangze Xu et al.

The utilization of the triplane-based radiance fields has gained attention in recent years due to its ability to effectively disentangle 3D scenes with a high-quality representation and low computation cost. A key requirement of this method is the precise input of camera poses. However, due to the local update property of the triplane, a similar joint estimation as previous joint pose-NeRF optimization works easily results in local minima. To this end, we propose the Disentangled Triplane Generation module to introduce global feature context and smoothness into triplane learning, which mitigates errors caused by local updating. Then, we propose the Disentangled Plane Aggregation to mitigate the entanglement caused by the common triplane feature aggregation during camera pose updating. In addition, we introduce a two-stage warm-start training strategy to reduce the implicit constraints caused by the triplane generator. Quantitative and qualitative results demonstrate that our proposed method achieves state-of-the-art performance in novel view synthesis with noisy or unknown camera poses, as well as efficient convergence of optimization. Project page: https://gaohchen.github.io/DiGARR/.

Jiachen Shen, Wenxuan Wang, Chen Chen et al.

The "pre-training then fine-tuning (FT)" paradigm is widely adopted to boost the model performance of deep learning-based methods for medical volumetric segmentation. However, conventional full FT incurs high computational and memory costs. Thus, it is of increasing importance to fine-tune pre-trained models for medical volumetric segmentation tasks in a both effective and parameter-efficient manner. In this paper, we introduce a new framework named Med-Tuning to realize parameter-efficient tuning (PET) for medical volumetric segmentation task and an efficient plug-and-play module named Med-Adapter for task-specific feature extraction. With a small number of tuned parameters, our framework enhances the 2D baselines's precision on segmentation tasks, which are pre-trained on natural images. Extensive experiments on three benchmark datasets (CT and MRI modalities) show that our method achieves better results than previous PET methods on volumetric segmentation tasks. Compared to full FT, Med-Tuning reduces the fine-tuned model parameters by up to 4x, with even better segmentation performance. Our project webpage is at \url{https://rubics-xuan.github.io/Med-Tuning/}.

Xiaoyun Zheng, Liwei Liao, Jianbo Jiao et al.

Self-supervised Object Segmentation (SOS) aims to segment objects without any annotations. Under conditions of multi-camera inputs, the structural, textural and geometrical consistency among each view can be leveraged to achieve fine-grained object segmentation. To make better use of the above information, we propose Surface representation based Self-supervised Object Segmentation (Surface-SOS), a new framework to segment objects for each view by 3D surface representation from multi-view images of a scene. To model high-quality geometry surfaces for complex scenes, we design a novel scene representation scheme, which decomposes the scene into two complementary neural representation modules respectively with a Signed Distance Function (SDF). Moreover, Surface-SOS is able to refine single-view segmentation with multi-view unlabeled images, by introducing coarse segmentation masks as additional input. To the best of our knowledge, Surface-SOS is the first self-supervised approach that leverages neural surface representation to break the dependence on large amounts of annotated data and strong constraints. These constraints typically involve observing target objects against a static background or relying on temporal supervision in videos. Extensive experiments on standard benchmarks including LLFF, CO3D, BlendedMVS, TUM and several real-world scenes show that Surface-SOS always yields finer object masks than its NeRF-based counterparts and surpasses supervised single-view baselines remarkably. Code is available at: https://github.com/zhengxyun/Surface-SOS.

Wenjin Wu, Aleš Leonardis, Linjiang Chen et al.

Infrared (IR) spectroscopy, a type of vibrational spectroscopy, is widely used for molecular structure determination and provides critical structural information for chemists. However, existing approaches for recovering molecular structures from IR spectra typically rely on one-dimensional SMILES strings or two-dimensional molecular graphs, which fail to capture the intricate relationship between spectral features and three-dimensional molecular geometry. Recent advances in diffusion models have greatly enhanced the ability to generate molecular structures in 3D space. Yet, no existing model has explored the distribution of 3D molecular geometries corresponding to a single IR spectrum. In this work, we introduce IR-GeoDiff, a latent diffusion model that recovers 3D molecular geometries from IR spectra by integrating spectral information into both node and edge representations of molecular structures. We evaluate IR-GeoDiff from both spectral and structural perspectives, demonstrating its ability to recover the molecular distribution corresponding to a given IR spectrum. Furthermore, an attention-based analysis reveals that the model is able to focus on characteristic functional group regions in IR spectra, qualitatively consistent with common chemical interpretation practices.

Dongheng Lin, Mengxue Qu, Kunyang Han et al.

Most video-anomaly research stops at frame-wise detection, offering little insight into why an event is abnormal, typically outputting only frame-wise anomaly scores without spatial or semantic context. Recent video anomaly localization and video anomaly understanding methods improve explainability but remain data-dependent and task-specific. We propose a unified reasoning framework that bridges the gap between temporal detection, spatial localization, and textual explanation. Our approach is built upon a chained test-time reasoning process that sequentially connects these tasks, enabling holistic zero-shot anomaly analysis without any additional training. Specifically, our approach leverages intra-task reasoning to refine temporal detections and inter-task chaining for spatial and semantic understanding, yielding improved interpretability and generalization in a fully zero-shot manner. Without any additional data or gradients, our method achieves state-of-the-art zero-shot performance across multiple video anomaly detection, localization, and explanation benchmarks. The results demonstrate that careful prompt design with task-wise chaining can unlock the reasoning power of foundation models, enabling practical, interpretable video anomaly analysis in a fully zero-shot manner. Project Page: https://rathgrith.github.io/Unified_Frame_VAA/.

Jiangbei Yue, Shuonan Yang, Tailin Chen et al.

Hateful videos present serious risks to online safety and real-world well-being, necessitating effective detection methods. Although multimodal classification approaches integrating information from several modalities outperform unimodal ones, they typically neglect that even minimal hateful content defines a video's category. Specifically, they generally treat all content uniformly, instead of emphasizing the hateful components. Additionally, existing multimodal methods cannot systematically capture structured information in videos, limiting the effectiveness of multimodal fusion. To address these limitations, we propose a novel multimodal dual-stream graph neural network model. It constructs an instance graph by separating the given video into several instances to extract instance-level features. Then, a complementary weight graph assigns importance weights to these features, highlighting hateful instances. Importance weights and instance features are combined to generate video labels. Our model employs a graph-based framework to systematically model structured relationships within and across modalities. Extensive experiments on public datasets show that our model is state-of-the-art in hateful video classification and has strong explainability. Code is available: https://github.com/Multimodal-Intelligence-Lab-MIL/MultiHateGNN.

Shuonan Yang, Tailin Chen, Rahul Singh et al.

The rapid proliferation of online multimedia content has intensified the spread of hate speech, presenting critical societal and regulatory challenges. While recent work has advanced multimodal hateful video detection, most approaches rely on coarse, video-level annotations that overlook the temporal granularity of hateful content. This introduces substantial label noise, as videos annotated as hateful often contain long non-hateful segments. In this paper, we investigate the impact of such label ambiguity through a fine-grained approach. Specifically, we trim hateful videos from the HateMM and MultiHateClip English datasets using annotated timestamps to isolate explicitly hateful segments. We then conduct an exploratory analysis of these trimmed segments to examine the distribution and characteristics of both hateful and non-hateful content. This analysis highlights the degree of semantic overlap and the confusion introduced by coarse, video-level annotations. Finally, controlled experiments demonstrated that time-stamp noise fundamentally alters model decision boundaries and weakens classification confidence, highlighting the inherent context dependency and temporal continuity of hate speech expression. Our findings provide new insights into the temporal dynamics of multimodal hateful videos and highlight the need for temporally aware models and benchmarks for improved robustness and interpretability. Code and data are available at https://github.com/Multimodal-Intelligence-Lab-MIL/HatefulVideoLabelNoise.

Bingfeng Zhang, Jimin Xiao, Jianbo Jiao et al.

Weakly supervised semantic segmentation is receiving great attention due to its low human annotation cost. In this paper, we aim to tackle bounding box supervised semantic segmentation, i.e., training accurate semantic segmentation models using bounding box annotations as supervision. To this end, we propose Affinity Attention Graph Neural Network ($A^2$GNN). Following previous practices, we first generate pseudo semantic-aware seeds, which are then formed into semantic graphs based on our newly proposed affinity Convolutional Neural Network (CNN). Then the built graphs are input to our $A^2$GNN, in which an affinity attention layer is designed to acquire the short- and long- distance information from soft graph edges to accurately propagate semantic labels from the confident seeds to the unlabeled pixels. However, to guarantee the precision of the seeds, we only adopt a limited number of confident pixel seed labels for $A^2$GNN, which may lead to insufficient supervision for training. To alleviate this issue, we further introduce a new loss function and a consistency-checking mechanism to leverage the bounding box constraint, so that more reliable guidance can be included for the model optimization. Experiments show that our approach achieves new state-of-the-art performances on Pascal VOC 2012 datasets (val: 76.5\%, test: 75.2\%). More importantly, our approach can be readily applied to bounding box supervised instance segmentation task or other weakly supervised semantic segmentation tasks, with state-of-the-art or comparable performance among almot all weakly supervised tasks on PASCAL VOC or COCO dataset. Our source code will be available at https://github.com/zbf1991/A2GNN.

Jiangliu Wang, Jianbo Jiao, Linchao Bao et al.

This paper proposes a novel pretext task to address the self-supervised video representation learning problem. Specifically, given an unlabeled video clip, we compute a series of spatio-temporal statistical summaries, such as the spatial location and dominant direction of the largest motion, the spatial location and dominant color of the largest color diversity along the temporal axis, etc. Then a neural network is built and trained to yield the statistical summaries given the video frames as inputs. In order to alleviate the learning difficulty, we employ several spatial partitioning patterns to encode rough spatial locations instead of exact spatial Cartesian coordinates. Our approach is inspired by the observation that human visual system is sensitive to rapidly changing contents in the visual field, and only needs impressions about rough spatial locations to understand the visual contents. To validate the effectiveness of the proposed approach, we conduct extensive experiments with four 3D backbone networks, i.e., C3D, 3D-ResNet, R(2+1)D and S3D-G. The results show that our approach outperforms the existing approaches across these backbone networks on four downstream video analysis tasks including action recognition, video retrieval, dynamic scene recognition, and action similarity labeling. The source code is publicly available at: https://github.com/laura-wang/video_repres_sts.

Jiangliu Wang, Jianbo Jiao, Yun-Hui Liu

This paper addresses the problem of self-supervised video representation learning from a new perspective -- by video pace prediction. It stems from the observation that human visual system is sensitive to video pace, e.g., slow motion, a widely used technique in film making. Specifically, given a video played in natural pace, we randomly sample training clips in different paces and ask a neural network to identify the pace for each video clip. The assumption here is that the network can only succeed in such a pace reasoning task when it understands the underlying video content and learns representative spatio-temporal features. In addition, we further introduce contrastive learning to push the model towards discriminating different paces by maximizing the agreement on similar video content. To validate the effectiveness of the proposed method, we conduct extensive experiments on action recognition and video retrieval tasks with several alternative network architectures. Experimental evaluations show that our approach achieves state-of-the-art performance for self-supervised video representation learning across different network architectures and different benchmarks. The code and pre-trained models are available at https://github.com/laura-wang/video-pace.

Richard Droste, Jianbo Jiao, J. Alison Noble

Visual saliency modeling for images and videos is treated as two independent tasks in recent computer vision literature. While image saliency modeling is a well-studied problem and progress on benchmarks like SALICON and MIT300 is slowing, video saliency models have shown rapid gains on the recent DHF1K benchmark. Here, we take a step back and ask: Can image and video saliency modeling be approached via a unified model, with mutual benefit? We identify different sources of domain shift between image and video saliency data and between different video saliency datasets as a key challenge for effective joint modelling. To address this we propose four novel domain adaptation techniques - Domain-Adaptive Priors, Domain-Adaptive Fusion, Domain-Adaptive Smoothing and Bypass-RNN - in addition to an improved formulation of learned Gaussian priors. We integrate these techniques into a simple and lightweight encoder-RNN-decoder-style network, UNISAL, and train it jointly with image and video saliency data. We evaluate our method on the video saliency datasets DHF1K, Hollywood-2 and UCF-Sports, and the image saliency datasets SALICON and MIT300. With one set of parameters, UNISAL achieves state-of-the-art performance on all video saliency datasets and is on par with the state-of-the-art for image saliency datasets, despite faster runtime and a 5 to 20-fold smaller model size compared to all competing deep methods. We provide retrospective analyses and ablation studies which confirm the importance of the domain shift modeling. The code is available at https://github.com/rdroste/unisal

Jiangliu Wang, Jianbo Jiao, Linchao Bao et al.

We address the problem of video representation learning without human-annotated labels. While previous efforts address the problem by designing novel self-supervised tasks using video data, the learned features are merely on a frame-by-frame basis, which are not applicable to many video analytic tasks where spatio-temporal features are prevailing. In this paper we propose a novel self-supervised approach to learn spatio-temporal features for video representation. Inspired by the success of two-stream approaches in video classification, we propose to learn visual features by regressing both motion and appearance statistics along spatial and temporal dimensions, given only the input video data. Specifically, we extract statistical concepts (fast-motion region and the corresponding dominant direction, spatio-temporal color diversity, dominant color, etc.) from simple patterns in both spatial and temporal domains. Unlike prior puzzles that are even hard for humans to solve, the proposed approach is consistent with human inherent visual habits and therefore easy to answer. We conduct extensive experiments with C3D to validate the effectiveness of our proposed approach. The experiments show that our approach can significantly improve the performance of C3D when applied to video classification tasks. Code is available at https://github.com/laura-wang/video_repres_mas.

Yuqian Zhou, Jianbo Jiao, Haibin Huang et al.

Discriminative learning-based image denoisers have achieved promising performance on synthetic noises such as Additive White Gaussian Noise (AWGN). The synthetic noises adopted in most previous work are pixel-independent, but real noises are mostly spatially/channel-correlated and spatially/channel-variant. This domain gap yields unsatisfied performance on images with real noises if the model is only trained with AWGN. In this paper, we propose a novel approach to boost the performance of a real image denoiser which is trained only with synthetic pixel-independent noise data dominated by AWGN. First, we train a deep model that consists of a noise estimator and a denoiser with mixed AWGN and Random Value Impulse Noise (RVIN). We then investigate Pixel-shuffle Down-sampling (PD) strategy to adapt the trained model to real noises. Extensive experiments demonstrate the effectiveness and generalization of the proposed approach. Notably, our method achieves state-of-the-art performance on real sRGB images in the DND benchmark among models trained with synthetic noises. Codes are available at https://github.com/yzhouas/PD-Denoising-pytorch.

Ding Liu, Bihan Wen, Jianbo Jiao et al.

Image denoising and high-level vision tasks are usually handled independently in the conventional practice of computer vision, and their connection is fragile. In this paper, we cope with the two jointly and explore the mutual influence between them with the focus on two questions, namely (1) how image denoising can help improving high-level vision tasks, and (2) how the semantic information from high-level vision tasks can be used to guide image denoising. First for image denoising we propose a convolutional neural network in which convolutions are conducted in various spatial resolutions via downsampling and upsampling operations in order to fuse and exploit contextual information on different scales. Second we propose a deep neural network solution that cascades two modules for image denoising and various high-level tasks, respectively, and use the joint loss for updating only the denoising network via back-propagation. We experimentally show that on one hand, the proposed denoiser has the generality to overcome the performance degradation of different high-level vision tasks. On the other hand, with the guidance of high-level vision information, the denoising network produces more visually appealing results. Extensive experiments demonstrate the benefit of exploiting image semantics simultaneously for image denoising and high-level vision tasks via deep learning. The code is available online: https://github.com/Ding-Liu/DeepDenoising

Kaiqiang Xiong, Rui Peng, Zhe Zhang et al.

Unsupervised Multi-View Stereo (MVS) methods have achieved promising progress recently. However, previous methods primarily depend on the photometric consistency assumption, which may suffer from two limitations: indistinguishable regions and view-dependent effects, e.g., low-textured areas and reflections. To address these issues, in this paper, we propose a new dual-level contrastive learning approach, named CL-MVSNet. Specifically, our model integrates two contrastive branches into an unsupervised MVS framework to construct additional supervisory signals. On the one hand, we present an image-level contrastive branch to guide the model to acquire more context awareness, thus leading to more complete depth estimation in indistinguishable regions. On the other hand, we exploit a scene-level contrastive branch to boost the representation ability, improving robustness to view-dependent effects. Moreover, to recover more accurate 3D geometry, we introduce an L0.5 photometric consistency loss, which encourages the model to focus more on accurate points while mitigating the gradient penalty of undesirable ones. Extensive experiments on DTU and Tanks&Temples benchmarks demonstrate that our approach achieves state-of-the-art performance among all end-to-end unsupervised MVS frameworks and outperforms its supervised counterpart by a considerable margin without fine-tuning.

Shuonan Yang, Tailin Chen, Jiangbei Yue et al.

Hate speech in online videos is posing an increasingly serious threat to digital platforms, especially as video content becomes increasingly multimodal and context-dependent. Existing methods often struggle to effectively fuse the complex semantic relationships between modalities and lack the ability to understand nuanced hateful content. To address these issues, we propose an innovative Reasoning-Aware Multimodal Fusion (RAMF) framework. To tackle the first challenge, we design Local-Global Context Fusion (LGCF) to capture both local salient cues and global temporal structures, and propose Semantic Cross Attention (SCA) to enable fine-grained multimodal semantic interaction. To tackle the second challenge, we introduce adversarial reasoning-a structured three-stage process where a vision-language model generates (i) objective descriptions, (ii) hate-assumed inferences, and (iii) non-hate-assumed inferences-providing complementary semantic perspectives that enrich the model's contextual understanding of nuanced hateful intent. Evaluations on two real-world hateful video datasets demonstrate that our method achieves robust generalisation performance, improving upon state-of-the-art methods by 3% and 7% in Macro-F1 and hate class recall, respectively. We will release the code after the anonymity period ends.

Yueming Sun, Long Yang, Jianbo Jiao et al.

The proliferation of hateful content in online videos poses severe threats to individual well-being and societal harmony. However, existing solutions for video hate detection either rely heavily on large-scale human annotations or lack fine-grained temporal precision. In this work, we propose LELA, the first training-free Large Language Model (LLM) based framework for hate video localization. Distinct from state-of-the-art models that depend on supervised pipelines, LELA leverages LLMs and modality-specific captioning to detect and temporally localize hateful content in a training-free manner. Our method decomposes a video into five modalities, including image, speech, OCR, music, and video context, and uses a multi-stage prompting scheme to compute fine-grained hateful scores for each frame. We further introduce a composition matching mechanism to enhance cross-modal reasoning. Experiments on two challenging benchmarks, HateMM and MultiHateClip, demonstrate that LELA outperforms all existing training-free baselines by a large margin. We also provide extensive ablations and qualitative visualizations, establishing LELA as a strong foundation for scalable and interpretable hate video localization.

Hao Chen, Yuqi Hou, Chenyuan Qu et al.

Human perception of the world is shaped by a multitude of viewpoints and modalities. While many existing datasets focus on scene understanding from a certain perspective (e.g. egocentric or third-person views), our dataset offers a panoptic perspective (i.e. multiple viewpoints with multiple data modalities). Specifically, we encapsulate third-person panoramic and front views, as well as egocentric monocular/binocular views with rich modalities including video, multi-channel audio, directional binaural delay, location data and textual scene descriptions within each scene captured, presenting comprehensive observation of the world. Figure 1 offers a glimpse of all 28 scene categories of our 360+x dataset. To the best of our knowledge, this is the first database that covers multiple viewpoints with multiple data modalities to mimic how daily information is accessed in the real world. Through our benchmark analysis, we presented 5 different scene understanding tasks on the proposed 360+x dataset to evaluate the impact and benefit of each data modality and perspective in panoptic scene understanding. We hope this unique dataset could broaden the scope of comprehensive scene understanding and encourage the community to approach these problems from more diverse perspectives.

Xiaoyun Zheng, Liwei Liao, Xufeng Li et al.

High-quality human reconstruction and photo-realistic rendering of a dynamic scene is a long-standing problem in computer vision and graphics. Despite considerable efforts invested in developing various capture systems and reconstruction algorithms, recent advancements still struggle with loose or oversized clothing and overly complex poses. In part, this is due to the challenges of acquiring high-quality human datasets. To facilitate the development of these fields, in this paper, we present PKU-DyMVHumans, a versatile human-centric dataset for high-fidelity reconstruction and rendering of dynamic human scenarios from dense multi-view videos. It comprises 8.2 million frames captured by more than 56 synchronized cameras across diverse scenarios. These sequences comprise 32 human subjects across 45 different scenarios, each with a high-detailed appearance and realistic human motion. Inspired by recent advancements in neural radiance field (NeRF)-based scene representations, we carefully set up an off-the-shelf framework that is easy to provide those state-of-the-art NeRF-based implementations and benchmark on PKU-DyMVHumans dataset. It is paving the way for various applications like fine-grained foreground/background decomposition, high-quality human reconstruction and photo-realistic novel view synthesis of a dynamic scene. Extensive studies are performed on the benchmark, demonstrating new observations and challenges that emerge from using such high-fidelity dynamic data.

Qiming Huang, Hao Ai, Jianbo Jiao

Benefiting from the inductive biases learned from large-scale datasets, open-vocabulary semantic segmentation (OVSS) leverages the power of vision-language models, such as CLIP, to achieve remarkable progress without requiring task-specific training. However, due to CLIP's pre-training nature on image-text pairs, it tends to focus on global semantic alignment, resulting in suboptimal performance when associating fine-grained visual regions with text. This leads to noisy and inconsistent predictions, particularly in local areas. We attribute this to a dispersed bias stemming from its contrastive training paradigm, which is difficult to alleviate using CLIP features alone. To address this, we propose a structure-aware feature rectification approach that incorporates instance-specific priors derived directly from the image. Specifically, we construct a region adjacency graph (RAG) based on low-level features (e.g., colour and texture) to capture local structural relationships and use it to refine CLIP features by enhancing local discrimination. Extensive experiments show that our method effectively suppresses segmentation noise, improves region-level consistency, and achieves strong performance on multiple open-vocabulary segmentation benchmarks.

Kai Fang, Anqi Zhang, Guangyu Gao et al.

Effective Class Incremental Segmentation (CIS) requires simultaneously mitigating catastrophic forgetting and ensuring sufficient plasticity to integrate new classes. The inherent conflict above often leads to a back-and-forth, which turns the objective into finding the balance between the performance of previous~(old) and incremental~(new) classes. To address this conflict, we introduce a novel approach, Conflict Mitigation via Branched Optimization~(CoMBO). Within this approach, we present the Query Conflict Reduction module, designed to explicitly refine queries for new classes through lightweight, class-specific adapters. This module provides an additional branch for the acquisition of new classes while preserving the original queries for distillation. Moreover, we develop two strategies to further mitigate the conflict following the branched structure, \textit{i.e.}, the Half-Learning Half-Distillation~(HDHL) over classification probabilities, and the Importance-Based Knowledge Distillation~(IKD) over query features. HDHL selectively engages in learning for classification probabilities of queries that match the ground truth of new classes, while aligning unmatched ones to the corresponding old probabilities, thus ensuring retention of old knowledge while absorbing new classes via learning negative samples. Meanwhile, IKD assesses the importance of queries based on their matching degree to old classes, prioritizing the distillation of important features and allowing less critical features to evolve. Extensive experiments in Class Incremental Panoptic and Semantic Segmentation settings have demonstrated the superior performance of CoMBO. Project page: https://guangyu-ryan.github.io/CoMBO.

Gensheng Pei, Yazhou Yao, Jianbo Jiao et al.

Conventional video object segmentation (VOS) methods usually necessitate a substantial volume of pixel-level annotated video data for fully supervised learning. In this paper, we present HVC, a \textbf{h}ybrid static-dynamic \textbf{v}isual \textbf{c}orrespondence framework for self-supervised VOS. HVC extracts pseudo-dynamic signals from static images, enabling an efficient and scalable VOS model. Our approach utilizes a minimalist fully-convolutional architecture to capture static-dynamic visual correspondence in image-cropped views. To achieve this objective, we present a unified self-supervised approach to learn visual representations of static-dynamic feature similarity. Firstly, we establish static correspondence by utilizing a priori coordinate information between cropped views to guide the formation of consistent static feature representations. Subsequently, we devise a concise convolutional layer to capture the forward / backward pseudo-dynamic signals between two views, serving as cues for dynamic representations. Finally, we propose a hybrid visual correspondence loss to learn joint static and dynamic consistency representations. Our approach, without bells and whistles, necessitates only one training session using static image data, significantly reducing memory consumption ($\sim$16GB) and training time ($\sim$\textbf{2h}). Moreover, HVC achieves state-of-the-art performance in several self-supervised VOS benchmarks and additional video label propagation tasks.

Kaiqiang Xiong, Ying Feng, Qi Zhang et al.

3D human reconstruction from a single image is a challenging problem and has been exclusively studied in the literature. Recently, some methods have resorted to diffusion models for guidance, optimizing a 3D representation via Score Distillation Sampling(SDS) or generating one back-view image for facilitating reconstruction. However, these methods tend to produce unsatisfactory artifacts (\textit{e.g.} flattened human structure or over-smoothing results caused by inconsistent priors from multiple views) and struggle with real-world generalization in the wild. In this work, we present \emph{MVD-HuGaS}, enabling free-view 3D human rendering from a single image via a multi-view human diffusion model. We first generate multi-view images from the single reference image with an enhanced multi-view diffusion model, which is well fine-tuned on high-quality 3D human datasets to incorporate 3D geometry priors and human structure priors. To infer accurate camera poses from the sparse generated multi-view images for reconstruction, an alignment module is introduced to facilitate joint optimization of 3D Gaussians and camera poses. Furthermore, we propose a depth-based Facial Distortion Mitigation module to refine the generated facial regions, thereby improving the overall fidelity of the reconstruction.Finally, leveraging the refined multi-view images, along with their accurate camera poses, MVD-HuGaS optimizes the 3D Gaussians of the target human for high-fidelity free-view renderings. Extensive experiments on Thuman2.0 and 2K2K datasets show that the proposed MVD-HuGaS achieves state-of-the-art performance on single-view 3D human rendering.

Qiyue Sun, Tailin Chen, Yinghui Zhang et al.

The rapid growth of video content on platforms such as TikTok and YouTube has intensified the spread of multimodal hate speech, where harmful cues emerge subtly and asynchronously across visual, acoustic, and textual streams. Existing research primarily focuses on video-level classification, leaving the practically crucial task of temporal localisation, identifying when hateful segments occur, largely unaddressed. This challenge is even more noticeable under weak supervision, where only video-level labels are available, and static fusion or classification-based architectures struggle to capture cross-modal and temporal dynamics. To address these challenges, we propose MultiHateLoc, the first framework designed for weakly-supervised multimodal hate localisation. MultiHateLoc incorporates (1) modality-aware temporal encoders to model heterogeneous sequential patterns, including a tailored text-based preprocessing module for feature enhancement; (2) dynamic cross-modal fusion to adaptively emphasise the most informative modality at each moment and a cross-modal contrastive alignment strategy to enhance multimodal feature consistency; (3) a modality-aware MIL objective to identify discriminative segments under video-level supervision. Despite relying solely on coarse labels, MultiHateLoc produces fine-grained, interpretable frame-level predictions. Experiments on HateMM and MultiHateClip show that our method achieves state-of-the-art performance in the localisation task.

Chenyuan Qu, Hao Chen, Jianbo Jiao

Exploring and understanding efficient image representations is a long-standing challenge in computer vision. While deep learning has achieved remarkable progress across image understanding tasks, its internal representations are often opaque, making it difficult to interpret how visual information is processed. In contrast, classical visual descriptors (e.g. edge, colour, and intensity distribution) have long been fundamental to image analysis and remain intuitively understandable to humans. Motivated by this gap, we ask a central question: Can modern learning benefit from these classical cues? In this paper, we answer it with VisualSplit, a framework that explicitly decomposes images into decoupled classical descriptors, treating each as an independent but complementary component of visual knowledge. Through a reconstruction-driven pre-training scheme, VisualSplit learns to capture the essence of each visual descriptor while preserving their interpretability. By explicitly decomposing visual attributes, our method inherently facilitates effective attribute control in various advanced visual tasks, including image generation and editing, extending beyond conventional classification and segmentation, suggesting the effectiveness of this new learning approach for visual understanding. Project page: https://chenyuanqu.com/VisualSplit/.

Haotian Li, Jianbo Jiao

Imagine living in a world composed solely of primitive shapes, could you still recognise familiar objects? Recent studies have shown that abstract images-constructed by primitive shapes-can indeed convey visual semantic information to deep learning models. However, representations obtained from such images often fall short compared to those derived from traditional raster images. In this paper, we study the reasons behind this performance gap and investigate how much high-level semantic content can be captured at different abstraction levels. To this end, we introduce the Hierarchical Abstraction Image Dataset (HAID), a novel data collection that comprises abstract images generated from normal raster images at multiple levels of abstraction. We then train and evaluate conventional vision systems on HAID across various tasks including classification, segmentation, and object detection, providing a comprehensive study between rasterised and abstract image representations. We also discuss if the abstract image can be considered as a potentially effective format for conveying visual semantic information and contributing to vision tasks.

Qiyue Sun, Qiming Huang, Yang Yang et al.

Humans usually show exceptional generalisation and discovery ability in the open world, when being shown uncommon new concepts. Whereas most existing studies in the literature focus on common typical data from closed sets, open-world novel discovery is under-explored in videos. In this paper, we are interested in asking: What if atypical unusual videos are exposed in the learning process? To this end, we collect a new video dataset consisting of various types of unusual atypical data (e.g., sci-fi, animation, etc.). To study how such atypical data may benefit open-world learning, we feed them into the model training process for representation learning. Focusing on three key tasks in open-world learning: out-of-distribution (OOD) detection, novel category discovery (NCD), and zero-shot action recognition (ZSAR), we found that even straightforward learning approaches with atypical data consistently improve performance across various settings. Furthermore, we found that increasing the categorical diversity of the atypical samples further boosts OOD detection performance. Additionally, in the NCD task, using a smaller yet more semantically diverse set of atypical samples leads to better performance compared to using a larger but more typical dataset. In the ZSAR setting, the semantic diversity of atypical videos helps the model generalise better to unseen action classes. These observations in our extensive experimental evaluations reveal the benefits of atypical videos for visual representation learning in the open world, together with the newly proposed dataset, encouraging further studies in this direction. The project page is at: https://julysun98.github.io/atypical_dataset.

Jiahao Wu, Rui Peng, Jianbo Jiao et al.

Due to the complex and highly dynamic motions in the real world, synthesizing dynamic videos from multi-view inputs for arbitrary viewpoints is challenging. Previous works based on neural radiance field or 3D Gaussian splatting are limited to modeling fine-scale motion, greatly restricting their application. In this paper, we introduce LocalDyGS, which consists of two parts to adapt our method to both large-scale and fine-scale motion scenes: 1) We decompose a complex dynamic scene into streamlined local spaces defined by seeds, enabling global modeling by capturing motion within each local space. 2) We decouple static and dynamic features for local space motion modeling. A static feature shared across time steps captures static information, while a dynamic residual field provides time-specific features. These are combined and decoded to generate Temporal Gaussians, modeling motion within each local space. As a result, we propose a novel dynamic scene reconstruction framework to model highly dynamic real-world scenes more realistically. Our method not only demonstrates competitive performance on various fine-scale datasets compared to state-of-the-art (SOTA) methods, but also represents the first attempt to model larger and more complex highly dynamic scenes. Project page: https://wujh2001.github.io/LocalDyGS/.

Dongheng Lin, Han Hu, Jianbo Jiao

Time becomes visible through illumination changes in what we see. Inspired by this, in this paper we explore the potential to learn time awareness from static images, trying to answer: *what time tells us?* To this end, we first introduce a Time-Oriented Collection (TOC) dataset, which contains 130,906 images with reliable timestamps. Leveraging this dataset, we propose a Time-Image Contrastive Learning (TICL) approach to jointly model timestamps and related visual representations through cross-modal contrastive learning. We found that the proposed TICL, 1) not only achieves state-of-the-art performance on the timestamp estimation task, over various benchmark metrics, 2) but also, interestingly, though only seeing static images, the time-aware embeddings learned from TICL show strong capability in several time-aware downstream tasks such as time-based image retrieval, video scene classification, and time-aware image editing. Our findings suggest that time-related visual cues can be learned from static images and are beneficial for various vision tasks, laying a foundation for future research on understanding time-related visual context. Project page: https://rathgrith.github.io/timetells_release/

Chen Xu, Qiming Huang, Yuqi Hou et al.

Medical image segmentation poses challenges due to domain gaps, data modality variations, and dependency on domain knowledge or experts, especially for low- and middle-income countries (LMICs). Whereas for humans, given a few exemplars (with corresponding labels), we are able to segment different medical images even without exten-sive domain-specific clinical training. In addition, current SAM-based medical segmentation models use fine-grained visual prompts, such as the bounding rectangle generated from manually annotated target segmentation mask, as the bounding box (bbox) prompt during the testing phase. However, in actual clinical scenarios, no such precise prior knowledge is available. Our experimental results also reveal that previous models nearly fail to predict when given coarser bbox prompts. Considering these issues, in this paper, we introduce a domain-aware selective adaptation approach to adapt the general knowledge learned from a large model trained with natural images to the corresponding medical domains/modalities, with access to only a few (e.g. less than 5) exemplars. Our method mitigates the aforementioned limitations, providing an efficient and LMICs-friendly solution. Extensive experimental analysis showcases the effectiveness of our approach, offering potential advancements in healthcare diagnostics and clinical applications in LMICs.

Zonglin Lyu, Ming Li, Jianbo Jiao et al.

Recent work in Video Frame Interpolation (VFI) tries to formulate VFI as a diffusion-based conditional image generation problem, synthesizing the intermediate frame given a random noise and neighboring frames. Due to the relatively high resolution of videos, Latent Diffusion Models (LDMs) are employed as the conditional generation model, where the autoencoder compresses images into latent representations for diffusion and then reconstructs images from these latent representations. Such a formulation poses a crucial challenge: VFI expects that the output is deterministically equal to the ground truth intermediate frame, but LDMs randomly generate a diverse set of different images when the model runs multiple times. The reason for the diverse generation is that the cumulative variance (variance accumulated at each step of generation) of generated latent representations in LDMs is large. This makes the sampling trajectory random, resulting in diverse rather than deterministic generations. To address this problem, we propose our unique solution: Frame Interpolation with Consecutive Brownian Bridge Diffusion. Specifically, we propose consecutive Brownian Bridge diffusion that takes a deterministic initial value as input, resulting in a much smaller cumulative variance of generated latent representations. Our experiments suggest that our method can improve together with the improvement of the autoencoder and achieve state-of-the-art performance in VFI, leaving strong potential for further enhancement.

Jianbo Jiao, João F. Henriques

In this work we investigate how to achieve equivariance to input transformations in deep networks, purely from data, without being given a model of those transformations. Convolutional Neural Networks (CNNs), for example, are equivariant to image translation, a transformation that can be easily modelled (by shifting the pixels vertically or horizontally). Other transformations, such as out-of-plane rotations, do not admit a simple analytic model. We propose an auto-encoder architecture whose embedding obeys an arbitrary set of equivariance relations simultaneously, such as translation, rotation, colour changes, and many others. This means that it can take an input image, and produce versions transformed by a given amount that were not observed before (e.g. a different point of view of the same object, or a colour variation). Despite extending to many (even non-geometric) transformations, our model reduces exactly to a CNN in the special case of translation-equivariance. Equivariances are important for the interpretability and robustness of deep networks, and we demonstrate results of successful re-rendering of transformed versions of input images on several synthetic and real datasets, as well as results on object pose estimation.

Chongjian Ge, Youwei Liang, Yibing Song et al.

Studies on self-supervised visual representation learning (SSL) improve encoder backbones to discriminate training samples without labels. While CNN encoders via SSL achieve comparable recognition performance to those via supervised learning, their network attention is under-explored for further improvement. Motivated by the transformers that explore visual attention effectively in recognition scenarios, we propose a CNN Attention REvitalization (CARE) framework to train attentive CNN encoders guided by transformers in SSL. The proposed CARE framework consists of a CNN stream (C-stream) and a transformer stream (T-stream), where each stream contains two branches. C-stream follows an existing SSL framework with two CNN encoders, two projectors, and a predictor. T-stream contains two transformers, two projectors, and a predictor. T-stream connects to CNN encoders and is in parallel to the remaining C-Stream. During training, we perform SSL in both streams simultaneously and use the T-stream output to supervise C-stream. The features from CNN encoders are modulated in T-stream for visual attention enhancement and become suitable for the SSL scenario. We use these modulated features to supervise C-stream for learning attentive CNN encoders. To this end, we revitalize CNN attention by using transformers as guidance. Experiments on several standard visual recognition benchmarks, including image classification, object detection, and semantic segmentation, show that the proposed CARE framework improves CNN encoder backbones to the state-of-the-art performance.

Zeyu Fu, Jianbo Jiao, Michael Suttie et al.

Fetal alcohol syndrome (FAS) caused by prenatal alcohol exposure can result in a series of cranio-facial anomalies, and behavioral and neurocognitive problems. Current diagnosis of FAS is typically done by identifying a set of facial characteristics, which are often obtained by manual examination. Anatomical landmark detection, which provides rich geometric information, is important to detect the presence of FAS associated facial anomalies. This imaging application is characterized by large variations in data appearance and limited availability of labeled data. Current deep learning-based heatmap regression methods designed for facial landmark detection in natural images assume availability of large datasets and are therefore not wellsuited for this application. To address this restriction, we develop a new regularized transfer learning approach that exploits the knowledge of a network learned on large facial recognition datasets. In contrast to standard transfer learning which focuses on adjusting the pre-trained weights, the proposed learning approach regularizes the model behavior. It explicitly reuses the rich visual semantics of a domain-similar source model on the target task data as an additional supervisory signal for regularizing landmark detection optimization. Specifically, we develop four regularization constraints for the proposed transfer learning, including constraining the feature outputs from classification and intermediate layers, as well as matching activation attention maps in both spatial and channel levels. Experimental evaluation on a collected clinical imaging dataset demonstrate that the proposed approach can effectively improve model generalizability under limited training samples, and is advantageous to other approaches in the literature.

Zeyu Fu, Jianbo Jiao, Michael Suttie et al.

Recently, there is an increasing demand for automatically detecting anatomical landmarks which provide rich structural information to facilitate subsequent medical image analysis. Current methods related to this task often leverage the power of deep neural networks, while a major challenge in fine tuning such models in medical applications arises from insufficient number of labeled samples. To address this, we propose to regularize the knowledge transfer across source and target tasks through cross-task representation learning. The proposed method is demonstrated for extracting facial anatomical landmarks which facilitate the diagnosis of fetal alcohol syndrome. The source and target tasks in this work are face recognition and landmark detection, respectively. The main idea of the proposed method is to retain the feature representations of the source model on the target task data, and to leverage them as an additional source of supervisory signals for regularizing the target model learning, thereby improving its performance under limited training samples. Concretely, we present two approaches for the proposed representation learning by constraining either final or intermediate model features on the target model. Experimental results on a clinical face image dataset demonstrate that the proposed approach works well with few labeled data, and outperforms other compared approaches.