Vision-centric semantic occupancy prediction plays a crucial role in autonomous driving, which requires accurate and reliable predictions from low-cost sensors. Although having notably narrowed the accuracy gap with LiDAR, there is still few research effort to explore the reliability and calibration in predicting semantic occupancy from camera. In this paper, we conduct a comprehensive evaluation of existing semantic occupancy prediction models from a reliability perspective for the first time. Despite the gradual alignment of camera-based models with LiDAR in terms of accuracy, a significant reliability gap still persists. To address this concern, we propose ReliOcc, a method designed to enhance the reliability of camera-based occupancy networks. ReliOcc provides a plug-and-play scheme for existing models, which integrates hybrid uncertainty from individual voxels with sampling-based noise and relative voxels through mix-up learning. Besides, an uncertainty-aware calibration strategy is devised to further improve model reliability in offline mode. Extensive experiments under various settings demonstrate that ReliOcc significantly enhances the reliability of learned model while maintaining the accuracy for both geometric and semantic predictions. Notably, our proposed approach exhibits robustness to sensor failures and out of domain noises during inference.
论文检索
输入标题、作者或关键词，从 1,280 篇学术成果中精准定位
Unsupervised domain adaptation (UDA) has been widely studied with the goal of transferring knowledge from a label-rich source domain to a related but unlabeled target domain. Most UDA techniques achieve this by reducing the feature discrepancies between the two domains to learn domain-invariant feature representations. While domain-invariant feature representations can reduce the differences between the source and target domains, excessively simplifying these differences may cause the model to overlook important domain-specific features, resulting in a decline in transfer learning effectiveness. To address this issue, this paper proposes a novel Gaussian Mixture Model for graph domain adaptation (GMM). This model effectively reduces the distributional bias between the source and target domains by modeling the distribution differences on a graph structure. GMM leverages the local structural information of the graph and the clustering capability of the Gaussian mixture model to automatically learn the latent mapping relationships between the source and target domains. To the best of our knowledge, this is the first work to introduce a Gaussian mixture model into UDA. Extensive experimental results on three standard benchmarks demonstrate that the proposed GMM algorithm outperforms state-of-the-art unsupervised domain adaptation methods in terms of performance.
Current general image forgery localization (GIFL) methods confront two main challenges: decoder overconffdence causing misidentiffcation of the authentic regions or incomplete predicted masks, and limited accuracy in localizing forgery details. Recently, diffusion models have excelled as dominant approach for generative models, particularly effective in capturing complex scene details. However, their potential for GIFL remains underexplored. Therefore, we propose a GIFL framework named ForgDiffuser with diffusion models. The core of ForgDiffuser lies in leveraging diffusion models conditioned on the forgery image to efffciently generate the segmentation mask for tampered regions. Speciffcally, we introduce the attentionguided module (AGM) to aggregate and enhance image feature representations. Meanwhile, we design the boundary-driven module (BDM) with edge supervision to improve the localization accuracy of boundary details. Additionally, the probabilistic modeling and stochastic sampling mechanisms of diffusion models effectively alleviate the overconffdence issue commonly observed in traditional decoders. Experiments on six benchmark datasets demonstrate that ForgDiffuser outperforms existing mainstream GIFL methods in both localization accuracy and robustness, especially under challenging manipulation conditions.
Diffusion model has underpinned much recent advances of dataset augmentation in various computer vision tasks. However, when involving generating multi-object images as real scenarios, most existing methods either rely entirely on text condition, resulting in a deviation between the generated objects and the original data, or rely too much on the original images, resulting in a lack of diversity in the generated images, which is of limited help to downstream tasks. To mitigate both problems with one stone, we propose a prompt-free conditional diffusion framework for multi-object image augmentation. Specifically, we introduce a local-global semantic fusion strategy to extract semantics from images to replace text, and inject knowledge into the diffusion model through LoRA to alleviate the category deviation between the original model and the target dataset. In addition, we design a reward model based counting loss to assist the traditional reconstruction loss for model training. By constraining the object counts of each category instead of pixel-by-pixel constraints, bridging the quantity deviation between the generated data and the original data while improving the diversity of the generated data. Experimental results demonstrate the superiority of the proposed method over several representative state-of-the-art baselines and showcase strong downstream task gain and out-of-domain generalization capabilities. Code is available at \href{https://github.com/00why00/PFCD}{here}.
BRIGHT-VO: Brightness-Guided Hybrid Transformer for Visual Odometry with Multi-modality Refinement Module
PDF ↗Visual odometry (VO) plays a crucial role in autonomous driving, robotic navigation, and other related tasks by estimating the position and orientation of a camera based on visual input. Significant progress has been made in data-driven VO methods, particularly those leveraging deep learning techniques to extract image features and estimate camera poses. However, these methods often struggle in low-light conditions because of the reduced visibility of features and the increased difficulty of matching keypoints. To address this limitation, we introduce BrightVO, a novel VO model based on Transformer architecture, which not only performs front-end visual feature extraction, but also incorporates a multi-modality refinement module in the back-end that integrates Inertial Measurement Unit (IMU) data. Using pose graph optimization, this module iteratively refines pose estimates to reduce errors and improve both accuracy and robustness. Furthermore, we create a synthetic low-light dataset, KiC4R, which includes a variety of lighting conditions to facilitate the training and evaluation of VO frameworks in challenging environments. Experimental results demonstrate that BrightVO achieves state-of-the-art performance on both the KiC4R dataset and the KITTI benchmarks. Specifically, it provides an average improvement of 20% in pose estimation accuracy in normal outdoor environments and 25% in low-light conditions, outperforming existing methods. This work is open-source at https://github.com/Anastasiawd/BrightVO.
Distinguishing key features in complex visual tasks is challenging. A novel approach treats image patches (tokens) as waves. By using both phase and amplitude, it captures richer semantics and specific invariances compared to pixel-based methods, and allows for feature fusion across regions for a holistic image representation. Based on this, we propose the Wave-wise Discriminative Transformer Tracker (WDT). During tracking, WDT represents features via phase-amplitude separation, enhancement, and mixture. First, we designed a Mutual Exclusive Phase-Amplitude Extractor (MEPAE) to separate phase and amplitude features with distinct semantics, representing spatial target info and background brightness respectively. Then, Wave-wise Feature Augmentation is carried out with two submodules: Phase-Amplitude Feature Augmentation and Mixture. The augmentation module disrupts the separated features in the same batch, and the mixture module recombines them to generate positive and negative waves. The original features are aggregated into the original wave. Positive waves have the same phase but different amplitudes, and negative waves have different phase components. Finally, self-supervised and tracking-supervised losses guide the global and local representation learning for original, positive, and negative waves, enhancing wave-level discrimination. Experiments on five benchmarks prove the effectiveness of our method.
Hand-drawn sketches are a natural and efficient medium for capturing and conveying ideas. Despite significant advancements in controllable natural image generation, translating freehand sketches into structured, machine-readable diagrams remains a labor-intensive and predominantly manual task. The primary challenge stems from the inherent ambiguity of sketches, which lack the structural constraints and semantic precision required for automated diagram generation. To address this challenge, we introduce SketchAgent, a multi-agent system designed to automate the transformation of hand-drawn sketches into structured diagrams. SketchAgent integrates sketch recognition, symbolic reasoning, and iterative validation to produce semantically coherent and structurally accurate diagrams, significantly reducing the need for manual effort. To evaluate the effectiveness of our approach, we propose the Sketch2Diagram Benchmark, a comprehensive dataset and evaluation framework encompassing eight diverse diagram categories, such as flowcharts, directed graphs, and model architectures. The dataset comprises over 6,000 high-quality examples with token-level annotations, standardized preprocessing, and rigorous quality control. By streamlining the diagram generation process, SketchAgent holds great promise for applications in design, education, and engineering, while offering a significant step toward bridging the gap between intuitive sketching and machine-readable diagram generation.
Automatically segmenting objects from optical remote sensing images (ORSIs) is an important task. Most existing models are primarily based on either convolutional or Transformer features, each offering distinct advantages. Exploiting both advantages is valuable research, but it presents several challenges, including the heterogeneity between the two types of features, high complexity, and large parameters of the model. However, these issues are often overlooked in existing the ORSIs methods, causing sub-optimal segmentation. For that, we propose a novel Dual-Perspective United Transformer (DPU-Former) with a unique structure designed to simultaneously integrate long-range dependencies and spatial details. In particular, we design the global-local mixed attention, which captures diverse information through two perspectives and introduces a Fourier-space merging strategy to obviate deviations for efficient fusion. Furthermore, we present a gated linear feed-forward network to increase the expressive ability. Additionally, we construct a DPU-Former decoder to aggregate and strength features at different layers. Consequently, the DPU-Former model outperforms the state-of-the-art methods on multiple datasets. Code: https://github.com/CSYSI/DPU-Former.
Understanding small visual objects is crucial in fields such as video surveillance, remote sensing, and autonomous driving. In this paper, we investigate the capability of advanced large vision-language models (LVLMs) to recognize and interpret small objects in visual data. To this end, we curate a specialized dataset for evaluating fine-grained visual hallucinations, incorporating two object categories and three types of hallucinations. First, we assess 11 state-of-the-art LVLMs, yielding several key insights, as anticipated, LVLMs perform significantly worse on queries related to small objects compared to regular-sized ones, with performance on regular objects proving to be an unreliable predictor of that on small objects. This finding underscores the need for dedicated research on fine-grained visual hallucinations. Second, we evaluate three training-free methods: Scaffold, Chain of Thought (CoT), and Image Resizing, all of which result in varying degrees of improvement. Furthermore, we conduct a series of detailed ablation studies on the visual encoders of Eagle-X5, examining their performance across fine-grained visual hallucination tasks. Our findings reveal that ConvNeXt architecture is critical for object existence recognition tasks. In contrast, for mitigating other types of hallucinations, integrating information from multiple visual encoders is significantly more effective than relying on a single encoder. These results highlight several promising directions for advancing small object recognition with LVLMs.
Audio-driven talking face synthesis (TFS) enables the creation of realistic speaking videos by combining a single facial image with a speech audio clip. Unlike other facial features that naturally deform during speech, teeth represent unique rigid structures whose shape and size should remain constant throughout the video sequence. However, current methods often produce temporal inconsistencies and artifacts in the teeth region, resulting in a less realistic appearance of the generated videos. To address this, we propose OrthoNet, a plug-and-play framework designed to eliminate unrealistic teeth effects in audio-driven TFS. Our method introduces a Detail-oriented Teeth Aligner module, designed to preserve teeth details and adapt to their shape. It works with a Memory-guided Teeth Stabilizer that integrates a long-term memory bank for global teeth structure and a short-term memory module for local temporal dynamics. Through this framework, OrthoNet acts like an orthodontist for existing Audio2Video methods, ensuring that teeth maintain natural rigidity and temporal consistency even under varying degrees of teeth occlusion. Extensive experiments demonstrate that our method makes the teeth in generated videos appear more natural during speech, significantly enhancing the temporal consistency and structural stability of audio-driven video generation.
Human pose estimation in low-light conditions is vital for applications such as surveillance and autonomous systems, yet the severe visual distortions hinder both manual annotation and estimation precision. Existing approaches typically rely on additional reference information to mitigate these issues, however, customized data collection equipment poses limitations on their scalability. To alleviate the issue, we construct a Low-Light Images and Poses (LLIP) dataset, which includes only paired low-light images and pose annotations obtained using off-the-shelf motion capture devices. Furthermore, we propose a Multi-grained High-frequency Feature Consistency Learning framework (MHFCL), which does not rely on additional reference information. MHFCL employs a Retinex-inspired restoration stream to recover high-frequency details and integrates them into pose estimation using a multi-grained consistency mechanism. Experiments demonstrate that our approach achieves a new benchmark in low-light pose estimation, while maintaining competitive performance in well-lit conditions.
CMFS: CLIP-Guided Modality Interaction for Mitigating Noise in Multi-Modal Image Fusion and Segmentation
PDF ↗Infrared-visible image fusion and semantic segmentation are pivotal tasks for robust scene understanding under challenging conditions such as low light. However, existing methods often struggle with high noise, modality inconsistencies, and inefficient cross-modal interactions, limiting fusion quality and segmentation accuracy. To this end, we propose CMFS, a unified framework that leverages CLIP-guided modality interaction to mitigate noise in multi-modal image fusion and segmentation. Our approach features a region-aware Modal Interaction Alignment module that combines a VMamba-based encoder with an additional shuffle layer to obtain more robust features and a CLIP-guided, regionally constrained multi-modal feature interaction block to emphasize foreground targets while suppressing low-light noise. Additionally, a Frequency-Spatial Collaboration module uses selective scanning and integrates wavelet-, spatial-, and Fourier-domain features to achieve adaptive denoising and balanced feature allocation. Furthermore, we employ a low-rank mixture-of-experts with dynamic routing to improve region-specific fusion and enhance pixel-level accuracy. Extensive experiments on several benchmarks show that, compared with state-of-the-art methods, the proposed approach demonstrates effectiveness in both image fusion quality and semantic segmentation accuracy, especially in complex environments. The source code will be released at IJCAI2025-CMFS.
While single-concept customization has been studied in 3D, multi-concept customization remains largely unexplored. To address this, we propose MultiDreamer3D that can generate coherent multi-concept 3D content in a divide-and-conquer manner. First, we generate 3D bounding boxes using an LLM-based layout controller. Next, a selective point cloud generator creates coarse point clouds for each concept. These point clouds are placed in the 3D bounding boxes and initialized into 3D Gaussian Splatting with concept labels, enabling precise identification of concept attributions in 2D projections. Finally, we refine 3D Gaussians via concept-aware interval score matching, guided by concept-aware diffusion. Our experimental results show that MultiDreamer3D not only ensures object presence and preserves the distinct identities of each concept but also successfully handles complex cases such as property change or interaction. To the best of our knowledge, we are the first to address the multi-concept customization in 3D.
MS-DPPs: Multi-Source Determinantal Point Processes for Contextual Diversity Refinement of Composite Attributes in Text to Image Retrieval
PDF ↗Result diversification (RD) is a crucial technique in Text-to-Image Retrieval for enhancing the efficiency of a practical application. Conventional methods focus solely on increasing the diversity metric of image appearances. However, the diversity metric and its desired value vary depending on the application, which limits the applications of RD. This paper proposes a novel task called CDR-CA (Contextual Diversity Refinement of Composite Attributes). CDR-CA aims to refine the diversities of multiple attributes, according to the application's context. To address this task, we propose Multi-Source DPPs, a simple yet strong baseline that extends the Determinantal Point Process (DPP) to multi-sources. We model MS-DPP as a single DPP model with a unified similarity matrix based on a manifold representation. We also introduce Tangent Normalization to reflect contexts. Extensive experiments demonstrate the effectiveness of the proposed method.
Modern multi-object tracking (MOT) predominantly relies on the tracking-by-detection paradigm to construct object trajectories. Traditional MOT attacks primarily degrade detection quality in specific frames only, lacking efficiency, while state-of-the-art (SOTA) approaches induce persistent identity (ID) switches by manipulating object positions during the association phase, even after the attack ends. In this paper, we reveal that these SOTA attacks can be easily counteracted by adjusting distance-related parameters in the association phase, exposing their lack of robustness. To overcome these limitations, we propose BankTweak, a novel adversarial attack targeting feature-based MOT systems to induce persistent ID switches (efficiency) without modifying object positions (robustness). BankTweak exploits a critical vulnerability in the Hungarian matching algorithm of MOT systems by strategically injecting altered features into feature banks during the association phase. Extensive experiments on MOT17 and MOT20 datasets, combining various detectors, feature extractors, and trackers, demonstrate that BankTweak significantly outperforms SOTA attacks up to 11.8 times, exposing fundamental vulnerabilities in the tracking-by-detection framework.
Partial point attack approaches focus on leveraging the fewest points to achieve the best attack efficiency for easy implementation in the physical domain. For the first time, this paper proposes that the partial point attack strategy should pay attention to not only the selection and disturbance of points, but also the penetration of current defense methods. By re-examining characteristics of previous partial point attack approaches leading to performance improvement, we discover two fundamental principles: first, the selection of attacked points should consider not only the favourable visual salience but also the proper position concentration, thus to acquire effective structural destruction on the basis of remaining imperceptible; second, the perturbation of target points should form meaningful structures rather than outliers. To achieve this, we first propose a novel distributed-concentrated point selection (DPS) strategy, which is easier to concentrate salient points containing rich local information in a few tiny regions. Additionally, to enhance the penetration efficacy and real-time performance of attack point clouds against defenses, we further design a perturbation network based on the multi-scale penetration loss (L_msp), which can generate adversarial samples with as few outliers as possible only through a single forward propagation. Experimental results demonstrate that the real-time distributed-concentrated penetration attack (RDPA) framework can achieve state-of-the-art (SOTA) success rates by perturbing only 3.5% of points, and have the best penetration for mainstream defense methods such as SRS and SOR.
Segment Anything Model 2 (SAM2) is a new-generation, high-precision model for image and video segmentation, offering extensive application prospects across numerous computer vision fields. However, as a large-scale model, its huge memory demands and expansive computing costs pose challenges for practical deployment. This paper presents Q-MiniSAM2, an efficient Quantization-based segmentation benchmark tailored to optimize SAM2 by Minimizing memory consumption and accelerating computations. We begin with applying Post-Training Quantization (PTQ) to SAM2, requiring only a relatively small dataset for network calibration, thereby eliminating the need for retraining. Building upon PTQ, we further introduce a Hierarchy-based Video Quantization method to enhance the model’s capacity to capture video semantics and temporal correlations across different time scales. Furthermore, we observe that SAM2’s memory overhead is predominantly concentrated on processing historical frames, and the redundant cross-attention computations significantly increase memory and computational costs due to the imperceptible change of the short time intervals between these frames. To tackle this issue, an Adaptive Mutual-KV mechanism is proposed to mitigate excessive cross-attention by leveraging inter-frame similarities. Comprehensive experiments demonstrate that the proposed approach achieves superior performance compared to state-of-the-art methods, underscoring its potential for efficient and scalable video segmentation.
Neural fields are now the central focus of research in 3D vision and computer graphics. Existing methods mainly focus on various scene representations, such as neural points and 3D Gaussians. However, few works have studied the rendering process to enhance the neural fields. In this work, we propose a plug-in method named K-Buffers that leverages multiple buffers to improve the rendering performance. Our method first renders K buffers from scene representations and constructs K pixel-wise feature maps. Then, We introduce a K-Feature Fusion Network (KFN) to merge the K pixel-wise feature maps. Finally, we adopt a feature decoder to generate the rendering image. We also introduce an acceleration strategy to improve rendering speed and quality. We apply our method to well-known radiance field baselines, including neural point fields and 3D Gaussian Splatting (3DGS). Extensive experiments demonstrate that our method effectively enhances the rendering performance of neural point fields and 3DGS.
ExpTalk: Diverse Emotional Expression via Adaptive Disentanglement and Refined Alignment for Speech-Driven 3D Facial Animation
PDF ↗Speech-driven 3D facial animation aims to create lifelike facial expressions that synchronize accurately with speech. Despite significant progress, many existing methods may focus on generating facial animation with a fixed emotional state, neglecting the diverse transformations of facial emotions under a given speech input. To solve this issue, we focus on exploring the refined alignment between speech representations and multiple domains in facial expression information. We aim to disentangle the spoken language and emotion facial priors from speech expressions, to guide the refinement of the facial vertices based on speech. To accomplish this objective, we propose ExpTalk, which first applies an Adaptive Disentanglement Variational Autoencoder (AD-VAE) to decouple facial expression aligned with spoken language and emotions of speech through contrastive learning. Then a Refined Alignment Diffusion (RAD) is employed to iteratively refine the decoupled facial expression priors through diffusion-based perturbations, producing facial animations that align with the emotional variations of the given speech. Extensive experiments prove the effectiveness of our ExpTalk by surpassing state-of-the-arts by a large margin.
Multimodal pathology-genomic analysis has become increasingly prominent in cancer survival prediction. However, existing studies mainly utilize multi-instance learning to aggregate patch-level features, neglecting the information loss of contextual and hierarchical details within pathology images. Furthermore, the disparity in data granularity and dimensionality between pathology and genomics leads to a significant modality imbalance. The high spatial resolution inherent in pathology data renders it a dominant role while overshadowing genomics in multimodal integration. In this paper, we propose a multimodal survival prediction framework that incorporates hypergraph learning to effectively capture both contextual and hierarchical details from pathology images. Moreover, it employs a modality rebalance mechanism and an interactive alignment fusion strategy to dynamically reweight the contributions of the two modalities, thereby mitigating the pathology-genomics imbalance. Quantitative and qualitative experiments are conducted on five TCGA datasets, demonstrating that our model outperforms advanced methods by over 3.4% in C-Index performance. Code: https://github.com/MCPathology/MRePath.