Current one-stage action detection methods, which simultaneously predict action boundaries and the corresponding class, do not estimate or use a measure of confidence in their boundary predictions, which can lead to inaccurate boundaries. We incorporate the estimation of boundary confidence into one-stage anchor-free detection, through an additional prediction head that predicts the refined boundaries with higher confidence. We obtain state-of-the-art performance on the challenging EPIC-KITCHENS-100 action detection as well as the standard THUMOS14 action detection benchmarks, and achieve improvement on the ActivityNet-1.3 benchmark.
Cite
@article{arxiv.2210.14284,
title = {Refining Action Boundaries for One-stage Detection},
author = {Hanyuan Wang and Majid Mirmehdi and Dima Damen and Toby Perrett},
journal= {arXiv preprint arXiv:2210.14284},
year = {2022}
}
Comments
Accepted to AVSS 2022. Our code is available at https://github.com/hanielwang/Refining_Boundary_Head.git