WACV 2026 论文列表

Event-based sensors offer high temporal resolution and low latency by generating sparse, asynchronous data. However, converting this irregular data into dense tensors for use in standard neural networks diminishes these inherent advantages, motivating research into graph representations. While such methods preserve sparsity and support asynchronous inference, their performance on downstream tasks remains limited due to suboptimal modeling of spatiotemporal dynamics. In this work, we propose a novel spatiotemporal multigraph representation to better capture spatial structure and temporal changes. Our approach constructs two decoupled graphs: a spatial graph leveraging B-spline basis functions to model global structure, and a temporal graph utilizing motion vector-based attention for local dynamic changes. This design enables the use of efficient 2D kernels in place of computationally expensive 3D kernels. We evaluate our method on the Gen1 automotive and eTraM datasets for event-based object detection, achieving over a 6% improvement in detection accuracy compared to previous graph-based works, with a 5xspeedup, reduced parameter count, and no increase in computational cost. These results highlight the effectiveness of structured graph modeling for asynchronous vision. Project page: https://eventbasedvision.github.io/eGSMV.

Extreme amodal detection is the task of inferring the 2D location of objects that are not fully visible in the input image but are visible within an expanded field-of-view. This differs from amodal detection, where the object is partially visible within the input image, but is occluded. In this paper, we consider the sub-problem of face detection, since this class provides motivating applications involving safety and privacy, but do not tailor our method specifically to this class. Existing approaches rely on image sequences so that missing detections may be interpolated from surrounding frames or make use of generative models to sample possible completions. In contrast, we consider the single-image task and propose a more efficient, sample-free approach that makes use of the contextual cues from the image to infer the presence of unseen faces. We design a heatmap-based extreme amodal object detector that addresses the problem of efficiently predicting a lot (the out-of-frame region) from a little (the image) with a selective coarse-to-fine decoder. Our method establishes strong results for this new task, even outperforming less efficient generative approaches. Code, data, and models are available at https://charliesong1999.github.io/exaft_web/.

Learning from (procedural) videos has increasingly served as a pathway for embodied agents to acquire skills from human demonstrations. To do this, video understanding models must be able to obtain structured understandings, such as the temporal segmentation of a demonstration into sequences of actions and skills, and to generalize the understandings to novel environments, tasks, and problem domains. In pursuit of this goal, we introduce Spacewalk-18, a benchmark containing two tasks: (1) step recognition and (2) video question answering, over a dataset of temporally segmented and labeled tasks in International Space Station spacewalk recordings. In tandem, the two tasks quantify a model's ability to: (1) generalize to novel domains; (2) utilize long temporal context and multimodal (e.g., visual and speech) information. Our extensive experimental analysis highlights the challenges of Spacewalk-18, but also suggests best practices for domain generalization and long-form understanding. Notably, we discover a promising adaptation via summarization technique that leads to significant performance improvement without model fine-tuning. The Spacewalk-18 benchmark is released at https://brown-palm.github.io/Spacewalk-18.

Active Speaker Detection (ASD) aims to identify who is speaking in complex visual scenes. While humans naturally rely on lip-audio synchronization, existing ASD models often misclassify non-speaking instances when lip movements and audio are unsynchronized. To address this, we propose Lip landmark Assisted Speaker dEtection for Robustness (LASER), which explicitly incorporates lip landmarks during training to guide the model's attention to speech-relevant regions. Given a face track, LASER extracts visual features and encodes 2D lip landmarks into dense maps. To handle failure cases such as low resolution or occlusion, we introduce an auxiliary consistency loss that aligns lip-aware and face-only predictions, removing the need for landmark detectors at test time. LASER outperforms state-of-the-art models across both in-domain and out-of-domain benchmarks. To further evaluate robustness in realistic conditions, we introduce LASER-bench, a curated dataset of modern video clips with varying levels of background noise. On the high-noise subset, LASER improves mAP by 3.3 and 4.3 points over LoCoNet and TalkNet, respectively, demonstrating strong resilience to real-world acoustic challenges.

Despite substantial progress in text-to-image generation, achieving precise text-image alignment remains challenging, particularly for prompts with rich compositional structure or imaginative elements. To address this, we introduce Negative Prompting for Image Correction (NPC), an automated pipeline that improves alignment by identifying and applying negative prompts that suppress unintended content. We begin by analyzing cross-attention patterns to explain why both targeted negatives--those directly tied to the prompt's alignment error--and untargeted negatives--tokens unrelated to the prompt but present in the generated image--can enhance alignment. To discover useful negatives, NPC generates candidate prompts using a verifier-captioner-proposer framework and ranks them with a salient text-space score, enabling effective selection without requiring additional image synthesis. On GenEval++ and Imagine-Bench, NPC outperforms strong baselines, achieving 0.571 vs. 0.371 on GenEval++ and the best overall performance on Imagine-Bench. By guiding what not to generate, NPC provides a principled, fully automated route to stronger text-image alignment in diffusion models. Code is released at https://github.com/wiarae/NPC.

High Dynamic Range (HDR) videos offer superior luminance and color fidelity as compared to Standard Dynamic Range (SDR) content. The rapid growth of User-Generated Content (UGC) on platforms such as YouTube, Instagram, and TikTok has brought a significant increase in the volumes of streamed and shared UGC videos. This newer category of videos brings new challenges to the development of effective No-Reference (NR) video quality assessment (VQA) models specialized to HDR UGC, because of the extreme variety and severities of distortions, arising from diverse capture, editing, and processing outcomes. Towards addressing this issue, we introduce BrightVQ, a sizeable new psychometric data resource. It is the first large-scale subjective video quality database dedicated to the quality modelling of HDR UGC videos. BrightVQ comprises 2,100 videos, on which we collected 73,794 perceptual quality ratings. Using this dataset, we also developed BrightRate, a novel video quality prediction model designed to capture both UGC-specific distortions coexisting with HDR-specific artifacts. Extensive experimental results demonstrate that BrightRate achieves state-of-the-art performance across HDR databases.

Hierarchical open-set classification handles previously unseen classes by assigning them to the most appropriate high-level category in a class taxonomy. We extend this paradigm to the semi-supervised setting, enabling the use of large-scale, uncurated datasets containing a mixture of known and unknown classes to improve the hierarchical open-set performance. To this end, we propose a teacher-student framework based on pseudo-labeling. Two key components are introduced: 1) subtree pseudo-labels, which provide reliable supervision in the presence of unknown data, and 2) age-gating, a mechanism that mitigates overconfidence in pseudo-labels. Experiments show that our framework outperforms self-supervised pretraining followed by supervised adaptation, and even matches the fully supervised counterpart when using only 20 labeled samples per class on the iNaturalist19 benchmark. Our code is available at https://github.com/walline/semihoc.

We present a novel approach for generating realistic speaking and talking faces by synthesizing a person's voice and facial movements from a static image, a voice profile, and a target text. The model encodes the prompt/driving text, the driving image, and the voice profile of an individual and then combines them to pass them to the multi-entangled latent space to foster key-value pairs and queries for the audio and video modality generation pipeline. The multi-entangled latent space is responsible for establishing the spatiotemporal person-specific features between the modalities. Further, entangled features are passed to the respective decoder of each modality for output audio and video generation. Our experiments and analysis through standard metrics demonstrate the effectiveness of our model. We also developed a domain-specific dataset for the problem. All model checkpoints, code, and the proposed dataset can be found at https://github.com/narratingForYou/NarratingForYou.

Vision-Language Models (VLMs) have shown capabilities in interpreting visual content, but their reliability in safety-critical scenarios remains insufficiently explored. We introduce VERI, a diagnostic benchmark comprising 200 synthetic images (100 contrastive pairs) and additional 50 real-world images (25 pairs) for validation. Each emergency scene is paired with a visually similar but safe counterpart through human verification. Using a two-stage evaluation protocol (risk identification and emergency response), we assess 17 VLMs across medical emergencies, accidents, and natural disasters. Our analysis reveals an "overreaction problem": models achieve high recall(70-100%) but suffer from low precision, misclassifying 31-96% of safe situations as dangerous. Seven safe scenarios were universally misclassified by all models. This "better-safe-than-sorry" bias stems from contextual overinterpretation (88-98% of errors). Both synthetic and real-world datasets confirm these systematic patterns, challenging VLM reliability in safety-critical applications. Addressing this requires enhanced contextual reasoning in ambiguous visual situations.

Identifying individual animals at different points in space and time is vital for effective wildlife monitoring and biodiversity conservation. While existing computer vision methods have shown promise in re-identifying animals, their capability in Animal Re-Identification (Animal ReID) remains restricted by the inherent visual variations, specifically high intra- and low inter-identity variations. High intra-identity variations refer to high visual diversity within the same individual due to pose or form changes and occlusions, and low inter-identity variations refer to subtle visual differences between distinct individuals due to fine-grained appearances. To address these challenges, we propose the Clip-based Animal RE-identification (CARE) framework, which leverages the image-conditioned textual description generation and individual-level semantic feature alignment, mitigating the negative impacts of visual variations in Animal ReID. Crucially, we have packaged CARE into a stand-alone toolkit and piloted it with stakeholders, facilitating real-world wildlife monitoring for biodiversity conservation. Extensive experiments on benchmark and in-the-wild datasets further demonstrate that CARE consistently outperforms state-of-the-art methods, validating its effectiveness in Animal ReID. Explore more about CARE at https://ml4sg.auckland.ac.nz/animal-re-identification-model/.

Recently, large vision-language models (LVLMs) have risen to be a promising approach for multimodal tasks. However, principled hallucination mitigation remains a critical challenge. In this work, we first analyze the data generation process in LVLM hallucination mitigation and affirm that on-policy data significantly outperforms off-policy data, which thus calls for efficient and reliable preference annotation of on-policy data. We then point out that, existing annotation methods introduce additional hallucination in training samples, which may enhance the model's hallucination patterns, to address this problem, we propose training a hallucination classifier giving binary annotations, which guarantee clean chosen samples for the subsequent alignment. To further harness of the power of on-policy data, we design a robust iterative direct preference optimization (DPO) algorithm adopting a dynamic sample reweighting scheme. We conduct comprehensive experiments on three benchmarks with comparison to 8 state-of-the-art baselines. In particular, our approach reduces the hallucination rate of LLaVA-1.5-7B on MMHalBench by 50.8% and the average hallucination rate on Object HalBench by 79.5%; more significantly, our method fully taps into the potential of open-source models, enabling LLaVA-1.5-13B to even surpass the performance of GPT-4V. Our code is available at https://github.com/MilchstraB/LVLM-OnPolicy.

Realistic signal generation and dataset augmentation are essential for advancing mmWave radar applications such as activity recognition and pose estimation, which rely heavily on diverse and environment-specific signal datasets. However, mmWave signals are inherently complex, sparse, and high-dimensional, making physical simulation computationally expensive. This paper presents mmWeaver, a novel framework that synthesizes realistic, environment-specific complex mmWave signals by modeling them as continuous functions using implicit neural representations (INRs), achieving up to 49-fold compression. mmWeaver incorporates hypernetworks that dynamically generate INR parameters based on environmental context (extracted from RGB-D images) and human motion features (derived from text-to-pose generation via MotionGPT), enabling efficient and adaptive signal synthesis. By conditioning on these semantic and geometric priors, mmWeaver generates diverse in-phase and quadrature (I/Q) signals at multiple resolutions, preserving information critical for downstream tasks such as point cloud estimation and activity classification. Extensive experiments show that mmWeaver achieves a complex structural similarity index (SSIM) of 0.88 and a peak signal-to-noise ratio (PSNR) of 35 dB, outperforming existing methods in signal realism while improving activity recognition accuracy by up to 7% and reducing human pose estimation error by up to 15%, all while operating 6-35 times faster than simulation-based approaches.

Humans excel at spatial-temporal reasoning, effortlessly interpreting dynamic visual events from an egocentric viewpoint. However, whether multimodal large language models (MLLMs) can similarly understand the 4D world remains uncertain. This paper explores multimodal spatial-temporal reasoning from an egocentric perspective, aiming to equip MLLMs with human-like reasoning capabilities. To support this objective, we introduce Ego-ST Bench, a novel benchmark containing over 5,000 question-answer pairs across four categories, systematically evaluating spatial, temporal, and integrated spatial-temporal reasoning. Additionally, we propose ST-R1 training paradigm, a video-based reasoning model that incorporates reverse thinking into its reinforcement learning process, significantly enhancing performance. We combine long-chain-of-thought (long-CoT) supervised fine-tuning with Group Relative Policy Optimization (GRPO) reinforcement learning, achieving notable improvements with limited high-quality data. Ego-ST Bench and ST-R1 provide valuable insights and resources for advancing video-based spatial-temporal reasoning research.

Automated ophthalmic report generation aims to reduce the diagnostic burden on retinal specialists by producing clinically accurate and standardized descriptions from medical imaging. However, current research predominantly remains fundus-centric and rarely exploits OCT-derived spatial evidence, limiting clinical transparency by obscuring which anatomical regions drive diagnostic decisions. To address these limitations, we propose LASOR (Lesion-Aware Segmentation-Guided Ophthalmic Report Generation), which extracts multi-scale features to robustly capture both small focal abnormalities and broader anatomical structures, generating reliable segmentation masks as spatial priors for report generation. Specifically, we utilize a lesion-aware patch weighting module to emphasize abnormal regions and leverage a curated instruction dataset incorporating spatial mask information to enhance the diagnostic capabilities of the proposed model. In addition, we introduce a mask-guided cross-modal consistency loss that strengthens vision-language alignment between pathological regions and their diagnostic descriptions. Extensive experiments on a retinal OCT dataset that includes twenty pathological conditions exhibit state-of-the-art performance, underscoring LASOR's potential to advance clinically transparent ophthalmic report generation systems.

Recent advances in 3D-aware generative models have enabled high-fidelity image synthesis of human identities. However, this progress raises urgent questions around user consent and the ability to remove specific individuals from a model's output space. We address this by introducing SUGAR, a framework for scalable generative unlearning that enables the removal of many identities (simultaneously or sequentially) without retraining the entire model. Rather than projecting unwanted identities to unrealistic outputs or relying on static template faces, SUGAR learns a personalized surrogate latent for each identity, diverting reconstructions to visually coherent alternatives while preserving the model's quality and diversity. We further introduce a continual utility preservation objective that guards against degradation as more identities are forgotten. SUGAR achieves state-of-the-art performance in removing up to 200 identities, while delivering up to a 700% improvement in retention utility compared to existing baselines. Our code is publicly available at https://github.com/judydnguyen/SUGARGenerative-Unlearn.

There is a growing demand for mobile user interface (UI) automation, driven by its broad applications across industries. With the advent of visual language models (VLMs), GUI automation has progressed from generating text-based instructions for humans to autonomously executing tasks, thus optimizing automation workflows. Recent approaches leverage VLMs for this problem due to their ability to 1) process on-screen content directly, 2) remain independent of device-specific APIs by utilizing human actions (e.g., clicks, typing), and 3) apply real-world contextual knowledge for task understanding. However, these models often have trouble accurately identifying widgets and determining actions due to limited spatial information in vision encoder features. Additionally, top-performing models are often large, requiring extensive training and resulting in inference delays. In this work, we introduce AFRAgent, an instruct-BLIP-based multimodal architecture that achieves superior performance in GUI automation while being less than one-fourth the size of its nearest competitor. To enhance image embeddings in the large language model (LLM) pipeline, we propose an adaptive feature renormalization-based (a token-level affine transformation) technique that effectively enriches low-resolution image embeddings and fuses high-resolution details. We evaluate AFRAgent on Meta-GUI and AITW benchmarks, establishing a new state-of-the-art baseline for smartphone automation.

Automated pallet load recognition is a critical task in industrial logistics, but the deployment of conventional deep learning systems is often unfeasible. Their reliance on large, manually annotated datasets creates a prohibitive bottleneck in terms of cost and time, especially in dynamic environments where product lines frequently change. To overcome this challenge, we introduce a highly flexible, dual-mode vision system built upon dense patch embeddings. Our primary, few-shot approach leverages features from the CAPI vision model to construct a compact memory bank from as little as a single labeled example per class. Classification is then performed via a simple yet highly effective k-nearest neighbor search. For annotation-free scenarios, we also propose a zero-shot mode that identifies the load by finding the rectangular region that minimizes intra-class feature variance. We demonstrate state-of-the-art performance on a new, challenging industrial dataset, where our few-shot method attains a mAP_ 50-95 over 90% with only one support image per class. Additionally, the fully unsupervised approach achieves a notable mAP_ 50-95 of up to 75%. The system's robustness and practical value were validated through its successful deployment in high-stakes, real-world scenarios. Our findings establish a basis for lightweight solutions that support the rapid, data-efficient integration of new vision systems into industrial workflows. To facilitate reproducibility, the source code is released at \normalsizehttps://github.com/juanjesus-ldo/F2ZLR .

We propose the Temporal Merge Adapter (TM-Adapter), a novel framework for image-to-video parameter-efficient transfer learning (PETL) specifically designed for temporal representation learning in video understanding. PETL has emerged as a practical strategy for adapting large-scale vision models to video tasks under limited computational budgets. However, existing PETL approaches suffer from local redundancy caused by highly similar consecutive frames, which limits the modeling of diverse temporal dependencies. To address this limitation, we introduce a lightweight merge-unmerge mechanism that temporally aggregates and redistributes token embeddings, enabling the model to capture diverse temporal patterns by mitigating redundancy. Furthermore, to effectively handle diverse temporal dependencies across different time scales, TM-Adapter introduces a single adapter module with two parallel branches, local and global adapters, each specialized in capturing complementary patterns at different temporal ranges. We validate our approach through experiments on Kinetics-400, Something-Something V2, and HMDB-51, demonstrating competitive performance compared to existing methods while maintaining high parameter efficiency.

Event cameras have gained significant attention due to their bio-inspired properties, such as high temporal resolution and high dynamic range. However, obtaining large-scale labeled ground-truth data for event-based vision tasks remains challenging and costly, which hinders the development of the event-based algorithm. In this paper, we present ControlEvents, a diffusion-based generative model designed to synthesize unlimited high-quality event data guided by diverse control signals such as class text labels, 2D skeletons, and 3D body poses. Our key insight is to leverage the diffusion prior from foundation models, such as Stable Diffusion, enabling high-quality event data generation with minimal fine-tuning and limited labeled data. Our method streamlines the data generation process and significantly reduces the cost of producing labeled event datasets. We demonstrate the effectiveness of our approach by synthesizing event data for visual recognition, 2D skeleton estimation, and 3D body pose estimation. Our experiments show that the synthesized labeled event data enhances model performance in all tasks. Additionally, our approach can generate events based on unseen text labels during training, illustrating the powerful text-based generation capabilities inherited from foundation models.