A Stronger Baseline for Ego-Centric Action Detection
This work provides an incremental improvement for researchers in computer vision focusing on action detection in ego-centric videos.
The paper tackles ego-centric action detection in untrimmed videos by separating classification and proposal tasks, achieving a 16.10% performance on the EPIC-KITCHENS-100 test set, which is an 11.7% improvement over the baseline.
This technical report analyzes an egocentric video action detection method we used in the 2021 EPIC-KITCHENS-100 competition hosted in CVPR2021 Workshop. The goal of our task is to locate the start time and the end time of the action in the long untrimmed video, and predict action category. We adopt sliding window strategy to generate proposals, which can better adapt to short-duration actions. In addition, we show that classification and proposals are conflict in the same network. The separation of the two tasks boost the detection performance with high efficiency. By simply employing these strategy, we achieved 16.10\% performance on the test set of EPIC-KITCHENS-100 Action Detection challenge using a single model, surpassing the baseline method by 11.7\% in terms of average mAP.