Boundary Discretization and Reliable Classification Network for Temporal Action Detection

TL;DR

This paper introduces BDRC-Net, a novel approach for temporal action detection that combines boundary discretization and reliable classification to improve accuracy and reduce false positives without relying on handcrafted anchors.

Contribution

The paper proposes BDRC-Net, which integrates boundary discretization and reliable classification modules to address limitations of mixed methods in temporal action detection.

Findings

Achieves competitive detection performance on benchmarks.

Effectively reduces false positives in predictions.

Eliminates need for handcrafted anchor design.

Abstract

Temporal action detection aims to recognize the action category and determine each action instance's starting and ending time in untrimmed videos. The mixed methods have achieved remarkable performance by seamlessly merging anchor-based and anchor-free approaches. Nonetheless, there are still two crucial issues within the mixed framework: (1) Brute-force merging and handcrafted anchor design hinder the substantial potential and practicality of the mixed methods. (2) Within-category predictions show a significant abundance of false positives. In this paper, we propose a novel Boundary Discretization and Reliable Classification Network (BDRC-Net) that addresses the issues above by introducing boundary discretization and reliable classification modules. Specifically, the boundary discretization module (BDM) elegantly merges anchor-based and anchor-free approaches in the form of boundary discretization, eliminating the need for the traditional handcrafted anchor design. Furthermore, the reliable classification module (RCM) predicts reliable global action categories to reduce false positives. Extensive experiments conducted on different benchmarks demonstrate that our proposed method achieves competitive detection performance. The code will be released at https://github.com/zhenyingfang/BDRC-Net.