Refining Action Boundaries for One-stage Detection

TL;DR

This paper introduces a boundary confidence estimation in one-stage action detection, improving boundary accuracy and achieving state-of-the-art results on multiple benchmarks.

Contribution

It proposes an additional prediction head for boundary confidence in one-stage anchor-free detection, enhancing boundary refinement and detection accuracy.

Findings

State-of-the-art performance on EPIC-KITCHENS-100

Improved results on THUMOS14

Enhanced boundary accuracy on ActivityNet-1.3

Abstract

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.