Uncertainty-Aware Mixture of Experts for Video Action Anticipation
文献类型：期刊论文
| 作者 | Qi, Zhaobo; Wang, Shuhui1,2,3; Zhang, Weigang2; Huang, Qingming3,4 |
| 刊名 | IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
 |
| 出版日期 | 2025-11-01 |
| 卷号 | 35期号:11页码:11158-11171 |
| 关键词 | Uncertainty Training Computational modeling Accuracy Transformers Parallel processing Integrated circuit modeling Benchmark testing Visualization Vectors Action anticipation video evolution uncertainty mixture of experts |
| ISSN号 | 1051-8215 |
| DOI | 10.1109/TCSVT.2025.3577027 |
| 英文摘要 | Anticipating future actions in daily life videos is crucial for seamless human-machine collaboration. However, accurately predicting these actions is challenging due to the inherent uncertainty and non-determinism of future events. To address this, we propose the uncertainty-aware mixture-of-experts framework for action anticipation (AntMoE), which employs multiple anticipation experts to model diverse video evolution patterns through learnable expert embeddings. These anticipation experts generate diverse predictions by integrating the top-k semantically similar observed video frames related to the current predicted feature representation, along with their corresponding expert embeddings. An anticipation router then aggregates these predictions based on the relationship between the current feature representation and all expert embeddings. To enhance the effectiveness of AntMoE, we introduce an expert regularization loss with three components: orthogonal loss promotes orthogonality among expert embeddings; expert balance loss ensures equal activation of all experts during training; and stability loss encourages the generation of numerically stable aggregation weights. Additionally, we incorporate an anticipation ranking loss function that aligns the model's confidence across varying anticipation time durations with the ground-truth ranking order, where a shorter anticipation time length corresponds to a higher confidence level. Experimental results across multiple benchmarks demonstrate that our method achieves remarkable anticipation performance. |
| 资助项目 | National Natural Science Foundation of China[62441232] ; National Natural Science Foundation of China[62306092] ; National Natural Science Foundation of China[U21B2038] ; National Natural Science Foundation of China[62476068] ; National Natural Science Foundation of China[62236008] ; Natural Science Foundation of Shandong Province, China[ZR2024QF066] |
| WOS研究方向 | Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:001607762600012 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/41600]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Zhang, Weigang |
| 作者单位 | 1.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 2.Harbin Inst Technol, Sch Comp Sci & Technol, Weihai 264209, Peoples R China 3.Peng Cheng Lab, Shenzhen 518066, Peoples R China 4.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 101408, Peoples R China |
| 推荐引用方式 GB/T 7714 | Qi, Zhaobo,Wang, Shuhui,Zhang, Weigang,et al. Uncertainty-Aware Mixture of Experts for Video Action Anticipation[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2025,35(11):11158-11171. |
| APA | Qi, Zhaobo,Wang, Shuhui,Zhang, Weigang,&Huang, Qingming.(2025).Uncertainty-Aware Mixture of Experts for Video Action Anticipation.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,35(11),11158-11171. |
| MLA | Qi, Zhaobo,et al."Uncertainty-Aware Mixture of Experts for Video Action Anticipation".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 35.11(2025):11158-11171. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。