Skip to Main content Skip to Navigation
Conference papers

ActionSpotter: Deep Reinforcement Learning Framework for Temporal Action Spotting in Videos

Abstract : Action spotting has recently been proposed as an alternative to action detection and key frame extraction. However, the current state-of-the-art method of action spotting requires an expensive ground truth composed of the search sequences employed by human annotators spotting actions - a critical limitation. In this article, we propose to use a reinforcement learning algorithm to perform efficient action spotting using only the temporal segments from the action detection annotations, thus opening an interesting solution for video understanding. Experiments performed on THUMOS14 and ActivityNet datasets show that the proposed method, named ActionSpotter, leads to good results and outperforms state-of-the-art detection outputs redrawn for this application. In particular, the spotting mean Average Precision on THUMOS14 is significantly improved from 59.7% to 65.6% while skipping 23% of video.
Complete list of metadata

Cited literature [51 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02534615
Contributor : Guillaume Vaudaux-Ruth Connect in order to contact the contributor
Submitted on : Thursday, November 5, 2020 - 11:43:01 AM
Last modification on : Sunday, June 26, 2022 - 2:55:35 AM

Files

bare_conf.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02534615, version 2

Citation

Guillaume Vaudaux-Ruth, Adrien Chan-Hon-Tong, Catherine Achard. ActionSpotter: Deep Reinforcement Learning Framework for Temporal Action Spotting in Videos. 2020 25th International Conference on Pattern Recognition (ICPR), Jan 2021, Milan, Italy. ⟨hal-02534615v2⟩

Share

Metrics

Record views

188

Files downloads

127