Two-Aspect Information Interaction Model for ABAW4 Multi-Task Challenge
Abstract
The task of ABAW is to predict frame-level emotion descriptors from videos: discrete emotional state; valence and arousal; and action units. In this paper, we propose the solution to the Multi-Task Learning (MTL) Challenge of the 4th Affective Behavior Analysis in-the-wild (ABAW) competition. Although researchers have proposed several approaches and achieved promising results in ABAW, current works in this task rarely consider interactions between different emotion descriptors. To this end, we propose a novel end to end architecture to achieve full integration of different types of information. Experimental results demonstrate the effectiveness of our proposed solution. Code are available at https://github.com/Swiftsss/ECCV2022MTL .
Cite
Text
Sun et al. "Two-Aspect Information Interaction Model for ABAW4 Multi-Task Challenge." European Conference on Computer Vision Workshops, 2022. doi:10.1007/978-3-031-25075-0_13Markdown
[Sun et al. "Two-Aspect Information Interaction Model for ABAW4 Multi-Task Challenge." European Conference on Computer Vision Workshops, 2022.](https://mlanthology.org/eccvw/2022/sun2022eccvw-twoaspect/) doi:10.1007/978-3-031-25075-0_13BibTeX
@inproceedings{sun2022eccvw-twoaspect,
title = {{Two-Aspect Information Interaction Model for ABAW4 Multi-Task Challenge}},
author = {Sun, Haiyang and Lian, Zheng and Liu, Bin and Tao, Jianhua and Sun, Licai and Cai, Cong and He, Yu},
booktitle = {European Conference on Computer Vision Workshops},
year = {2022},
pages = {173-180},
doi = {10.1007/978-3-031-25075-0_13},
url = {https://mlanthology.org/eccvw/2022/sun2022eccvw-twoaspect/}
}