Uncertainty-aware Action Decoupling Transformer for Action Anticipation

H Guo, N Agarwal, SY Lo, K Lee… - Proceedings of the IEEE …, 2024 - openaccess.thecvf.com
Human action anticipation aims at predicting what people will do in the future based on past
observations. In this paper we introduce Uncertainty-aware Action Decoupling Transformer …

Rethinking clip-based video learners in cross-domain open-vocabulary action recognition

KY Lin, H Ding, J Zhou, YM Tang, YX Peng… - arXiv preprint arXiv …, 2024 - arxiv.org
Building upon the impressive success of CLIP (Contrastive Language-Image Pretraining),
recent pioneer works have proposed to adapt the powerful CLIP to video data, leading to …

Divide and Fuse: Body Part Mesh Recovery from Partially Visible Human Images

T Luan, Z Gao, L Xie, A Sharma, H Ding… - … on Computer Vision, 2025 - Springer
We introduce a novel bottom-up approach for human body mesh reconstruction, specifically
designed to address the challenges posed by partial visibility and occlusion in input images …

SMART-vision: survey of modern action recognition techniques in vision

AK AlShami, R Rabinowitz, K Lam, Y Shleibik… - Multimedia Tools and …, 2024 - Springer
Abstract Human Action Recognition (HAR) is a challenging domain in computer vision,
involving recognizing complex patterns by analyzing the spatiotemporal dynamics of …

Trusted Video-Based Sewer Inspection via Support Clip-Based Pareto-Optimal Evidential Network

C Zhao, C Hu, H Shao, F Dunkin… - IEEE Signal Processing …, 2024 - ieeexplore.ieee.org
An automatic vision-based sewer inspection plays a vital role of sewage system in a modern
city. Existing methods have utilized evidential deep learning to construct trusted models …

Scene adaptive mechanism for action recognition

C Wu, XJ Wu, T Xu, J Kittler - Computer Vision and Image Understanding, 2024 - Elsevier
Scene knowledge plays an important role in visual analysis. For the task of action
recognition, human activities often occur in specific scenes. However, it should be …

Open-Set Biometrics: Beyond Good Closed-Set Models

Y Su, M Kim, F Liu, A Jain, X Liu - European Conference on Computer …, 2025 - Springer
Biometric recognition has primarily addressed closed-set identification, assuming all probe
subjects are in the gallery. However, most practical applications involve open-set biometrics …

CASR: Refining Action Segmentation via Magrinalizing Frame-levle Causal Relationships

K Du, X Yang, H Chen - arXiv preprint arXiv:2311.12401, 2023 - arxiv.org
Integrating deep learning and causal discovery has increased the interpretability of
Temporal Action Segmentation (TAS) tasks. However, frame-level causal relationships exist …

Fairness and Bias Mitigation in Computer Vision: A Survey

S Dehdashtian, R He, Y Li, G Balakrishnan… - arXiv preprint arXiv …, 2024 - arxiv.org
Computer vision systems have witnessed rapid progress over the past two decades due to
multiple advances in the field. As these systems are increasingly being deployed in high …

ContextHOI: Spatial Context Learning for Human-Object Interaction Detection

M Jia, L Zhao, G Li, Y Zheng - arXiv preprint arXiv:2412.09050, 2024 - arxiv.org
Spatial contexts, such as the backgrounds and surroundings, are considered critical in
Human-Object Interaction (HOI) recognition, especially when the instance-centric …