This technical report introduces our 2nd place solution to Kinetics-TPS Track on Part-level Action Parsing in ICCV DeeperAction Workshop 2021. Our entry is mainly based on YOLOF for instance and part detection, HRNet for human pose estimation, and CSN for video-level action recognition and frame-level part state parsing. We describe technical details for the Kinetics-TPS dataset, together with some experimental results. In the competition, we achieved 61.37% mAP on the test set of Kinetics-TPS.
@article{arxiv.2110.03368,
title = {A Baseline Framework for Part-level Action Parsing and Action Recognition},
author = {Xiaodong Chen and Xinchen Liu and Kun Liu and Wu Liu and Tao Mei},
journal= {arXiv preprint arXiv:2110.03368},
year = {2022}
}
Comments
4 pages, 1 figures, ICCV 2021 Challenge, 2nd place solution for Kinetics-TPS Track in ICCV DeeperAction Workshop 2021