From Pixels to Privacy: Temporally Consistent Video Anonymization via Token Pruning for Privacy Preserving Action Recognition

Recent advances in large-scale video models have significantly improved video understanding across domains such as surveillance, healthcare, and entertainment. However, these models also amplify privacy risks by encoding sensitive attributes, including facial identity, race, and gender. While image anonymization has been extensively studied, video anonymization remains relatively underexplored, even though modern video models can leverage spatiotemporal motion patterns as biometric identifiers. To address this challenge, we propose a novel attention-driven spatiotemporal video anonymization framework based on systematic disentanglement of utility and privacy features. Our key insight is that attention mechanisms in Vision Transformers (ViTs) can be explicitly structured to separate action-relevant information from privacy-sensitive content. Building on this insight, we introduce two task-specific classification tokens, an action CLS token and a privacy CLS token, that learn complementary representations within a shared Transformer backbone. We contrast their attention distributions to compute a utility-privacy score for each spatiotemporal tubelet, and keep the top-k tubelets with the highest scores. This selectively prunes tubelets dominated by privacy cues while preserving those most critical for action recognition. Extensive experiments demonstrate that our approach maintains action recognition performance comparable to models trained on raw videos, while substantially reducing privacy leakage. These results indicate that attention-driven spatiotemporal pruning offers an effective and principled solution for privacy-preserving video analytics.

PerceptionComp: A Video Benchmark for Complex Perception-Centric Reasoning

We introduce PerceptionComp, a manually annotated benchmark for complex, long-horizon, perception-centric video reasoning. PerceptionComp is designed so that no single moment is sufficient: answering each question requires multiple temporally separated pieces of visual evidence and compositional constraints under conjunctive and sequential logic, spanning perceptual subtasks such as objects, attributes, relations, locations, actions, and events, and requiring skills including semantic recognition, visual correspondence, temporal reasoning, and spatial reasoning. The benchmark contains 1,114 highly complex questions on 279 videos from diverse domains including city walk tours, indoor villa tours, video games, and extreme outdoor sports, with 100% manual annotation. Human studies show that PerceptionComp requires substantial test-time thinking and repeated perception steps: participants take much longer than on prior benchmarks, and accuracy drops to near chance (18.97%) when rewatching is disallowed. State-of-the-art MLLMs also perform substantially worse on PerceptionComp than on existing benchmarks: the best model in our evaluation, Gemini-3-Flash, reaches only 45.96% accuracy in the five-choice setting, while open-source models remain below 40%. These results suggest that perception-centric long-horizon video reasoning remains a major bottleneck, and we hope PerceptionComp will help drive progress in perceptual reasoning.

MA-Bench: Towards Fine-grained Micro-Action Understanding

With the rapid development of Multimodal Large Language Models (MLLMs), their potential in Micro-Action understanding, a vital role in human emotion analysis, remains unexplored due to the absence of specialized benchmarks. To tackle this issue, we present MA-Bench, a benchmark comprising 1,000 videos and a three-tier evaluation architecture that progressively examines micro-action perception, relational comprehension, and interpretive reasoning. MA-Bench contains 12,000 structured question-answer pairs, enabling systematic assessment of both recognition accuracy and action interpretation. The results of 23 representative MLLMs reveal that there are significant challenges in capturing motion granularity and fine-grained body-part dynamics. To address these challenges, we further construct MA-Bench-Train, a large-scale training corpus with 20.5K videos annotated with structured micro-action captions for fine-tuning MLLMs. The results of Qwen3-VL-8B fine-tuned on MA-Bench-Train show clear performance improvements across micro-action reasoning and explanation tasks. Our work aims to establish a foundation benchmark for advancing MLLMs in understanding subtle micro-action and human-related behaviors. Project Page: https://MA-Bench.github.io

Knowledge is Power: Advancing Few-shot Action Recognition with Multimodal Semantics from MLLMs

Multimodal Large Language Models (MLLMs) have propelled the field of few-shot action recognition (FSAR). However, preliminary explorations in this area primarily focus on generating captions to form a suboptimal feature->caption->feature pipeline and adopt metric learning solely within the visual space. In this paper, we propose FSAR-LLaVA, the first end-to-end method to leverage MLLMs (such as Video-LLaVA) as a multimodal knowledge base for directly enhancing FSAR. First, at the feature level, we leverage the MLLM's multimodal decoder to extract spatiotemporally and semantically enriched representations, which are then decoupled and enhanced by our Multimodal Feature-Enhanced Module into distinct visual and textual features that fully exploit their semantic knowledge for FSAR. Next, we leverage the versatility of MLLMs to craft input prompts that flexibly adapt to diverse scenarios, and use their aligned outputs to drive our designed Composite Task-Oriented Prototype Construction, effectively bridging the distribution gap between meta-train and meta-test sets. Finally, to enable multimodal features to guide metric learning jointly, we introduce a training-free Multimodal Prototype Matching Metric that adaptively selects the most decisive cues and efficiently leverages the decoupled feature representations produced by MLLMs. Extensive experiments demonstrate superior performance across various tasks with minimal trainable parameters.

CHIRP dataset: towards long-term, individual-level, behavioral monitoring of bird populations in the wild

Long-term behavioral monitoring of individual animals is crucial for studying behavioral changes that occur over different time scales, especially for conservation and evolutionary biology. Computer vision methods have proven to benefit biodiversity monitoring, but automated behavior monitoring in wild populations remains challenging. This stems from the lack of datasets that cover a range of computer vision tasks necessary to extract biologically meaningful measurements of individual animals. Here, we introduce such a dataset (CHIRP) with a new method (CORVID) for individual re-identification of wild birds. The CHIRP (Combining beHaviour, Individual Re-identification and Postures) dataset is curated from a long-term population of wild Siberian jays studied in Swedish Lapland, supporting re-identification (re-id), action recognition, 2D keypoint estimation, object detection, and instance segmentation. In addition to traditional task-specific benchmarking, we introduce application-specific benchmarking with biologically relevant metrics (feeding rates, co-occurrence rates) to evaluate the performance of models in real-world use cases. Finally, we present CORVID (COlouR-based Video re-ID), a novel pipeline for individual identification of birds based on the segmentation and classification of colored leg rings, a widespread approach for visual identification of individual birds. CORVID offers a probability-based id tracking method by matching the detected combination of color rings with a database. We use application-specific benchmarking to show that CORVID outperforms state-of-the-art re-id methods. We hope this work offers the community a blueprint for curating real-world datasets from ethically approved biological studies to bridge the gap between computer vision research and biological applications.

AG-EgoPose: Leveraging Action-Guided Motion and Kinematic Joint Encoding for Egocentric 3D Pose Estimation

Egocentric 3D human pose estimation remains challenging due to severe perspective distortion, limited body visibility, and complex camera motion inherent in first-person viewpoints. Existing methods typically rely on single-frame analysis or limited temporal fusion, which fails to effectively leverage the rich motion context available in egocentric videos. We introduce AG-EgoPose, a novel dual-stream framework that integrates short- and long-range motion context with fine-grained spatial cues for robust pose estimation from fisheye camera input. Our framework features two parallel streams: A spatial stream uses a weight-sharing ResNet-18 encoder-decoder to generate 2D joint heatmaps and corresponding joint-specific spatial feature tokens. Simultaneously, a temporal stream uses a ResNet-50 backbone to extract visual features, which are then processed by an action recognition backbone to capture the motion dynamics. These complementary representations are fused and refined in a transformer decoder with learnable joint tokens, which allows for the joint-level integration of spatial and temporal evidence while maintaining anatomical constraints. Experiments on real-world datasets demonstrate that AG-EgoPose achieves state-of-the-art performance in both quantitative and qualitative metrics. Code is available at: https://github.com/Mushfiq5647/AG-EgoPose.

Privacy-Preserving Federated Action Recognition via Differentially Private Selective Tuning and Efficient Communication

Federated video action recognition enables collaborative model training without sharing raw video data, yet remains vulnerable to two key challenges: \textit{model exposure} and \textit{communication overhead}. Gradients exchanged between clients and the server can leak private motion patterns, while full-model synchronization of high-dimensional video networks causes significant bandwidth and communication costs. To address these issues, we propose \textit{Federated Differential Privacy with Selective Tuning and Efficient Communication for Action Recognition}, namely \textit{FedDP-STECAR}. Our \textit{FedDP-STECAR} framework selectively fine-tunes and perturbs only a small subset of task-relevant layers under Differential Privacy (DP), reducing the surface of information leakage while preserving temporal coherence in video features. By transmitting only the tuned layers during aggregation, communication traffic is reduced by over 99\% compared to full-model updates. Experiments on the UCF-101 dataset using the MViT-B-16x4 transformer show that \textit{FedDP-STECAR} achieves up to \textbf{70.2\% higher accuracy} under strict privacy ($ε=0.65$) in centralized settings and \textbf{48\% faster training} with \textbf{73.1\% accuracy} in federated setups, enabling scalable and privacy-preserving video action recognition. Code available at https://github.com/izakariyya/mvit-federated-videodp

Unified Spatio-Temporal Token Scoring for Efficient Video VLMs

Token pruning is essential for enhancing the computational efficiency of vision-language models (VLMs), particularly for video-based tasks where temporal redundancy is prevalent. Prior approaches typically prune tokens either (1) within the vision transformer (ViT) exclusively for unimodal perception tasks such as action recognition and object segmentation, without adapting to downstream vision-language tasks; or (2) only within the LLM while leaving the ViT output intact, often requiring complex text-conditioned token selection mechanisms. In this paper, we introduce Spatio-Temporal Token Scoring (STTS), a simple and lightweight module that prunes vision tokens across both the ViT and the LLM without text conditioning or token merging, and is fully compatible with end-to-end training. By learning how to score temporally via an auxiliary loss and spatially via LLM downstream gradients, aided by our efficient packing algorithm, STTS prunes 50% of vision tokens throughout the entire architecture, resulting in a 62% improvement in efficiency during both training and inference with only a 0.7% drop in average performance across 13 short and long video QA tasks. Efficiency gains increase with more sampled frames per video. Applying test-time scaling for long-video QA further yields performance gains of 0.5-1% compared to the baseline. Overall, STTS represents a novel, simple yet effective technique for unified, architecture-wide vision token pruning.

SAW: Toward a Surgical Action World Model via Controllable and Scalable Video Generation

A surgical world model capable of generating realistic surgical action videos with precise control over tool-tissue interactions can address fundamental challenges in surgical AI and simulation -- from data scarcity and rare event synthesis to bridging the sim-to-real gap for surgical automation. However, current video generation methods, the very core of such surgical world models, require expensive annotations or complex structured intermediates as conditioning signals at inference, limiting their scalability. Other approaches exhibit limited temporal consistency across complex laparoscopic scenes and do not possess sufficient realism. We propose Surgical Action World (SAW) -- a step toward surgical action world modeling through video diffusion conditioned on four lightweight signals: language prompts encoding tool-action context, a reference surgical scene, tissue affordance mask, and 2D tool-tip trajectories. We design a conditional video diffusion approach that reformulates video-to-video diffusion into trajectory-conditioned surgical action synthesis. The backbone diffusion model is fine-tuned on a custom-curated dataset of 12,044 laparoscopic clips with lightweight spatiotemporal conditioning signals, leveraging a depth consistency loss to enforce geometric plausibility without requiring depth at inference. SAW achieves state-of-the-art temporal consistency (CD-FVD: 199.19 vs. 546.82) and strong visual quality on held-out test data. Furthermore, we demonstrate its downstream utility for (a) surgical AI, where augmenting rare actions with SAW-generated videos improves action recognition (clipping F1-score: 20.93% to 43.14%; cutting: 0.00% to 8.33%) on real test data, and (b) surgical simulation, where rendering tool-tissue interaction videos from simulator-derived trajectory points toward a visually faithful simulation engine.

Solution for 10th Competition on Ambivalence/Hesitancy (AH) Video Recognition Challenge using Divergence-Based Multimodal Fusion

We address the Ambivalence/Hesitancy (A/H) Video Recognition Challenge at the 10th ABAW Competition (CVPR 2026). We propose a divergence-based multimodal fusion that explicitly measures cross-modal conflict between visual, audio, and textual channels. Visual features are encoded as Action Units (AUs) extracted via Py-Feat, audio via Wav2Vec 2.0, and text via BERT. Each modality is processed by a BiLSTM with attention pooling and projected into a shared embedding space. The fusion module computes pairwise absolute differences between modality embeddings, directly capturing the incongruence that characterizes A/H. On the BAH dataset, our approach achieves a Macro F1 of 0.6808 on the validation test set, outperforming the challenge baseline of 0.2827. Statistical analysis across 1{,}132 videos confirms that temporal variability of AUs is the dominant visual discriminator of A/H.