Submission to Generic Event Boundary Detection Challenge@CVPR 2022: Local Context Modeling and Global Boundary Decoding Approach

TL;DR

This paper introduces a local context modeling and global boundary decoding approach for generic event boundary detection in videos, significantly improving accuracy by over 22% F1-score on the Kinetics-GEBD dataset.

Contribution

It proposes a novel combination of local context modeling and global boundary decoding for more accurate event boundary detection in videos.

Findings

Achieved 85.13% F1-score on Kinetics-GEBD dataset.

Improved F1-score by over 22% compared to baseline.

Demonstrated effectiveness of combined local and global modeling.

Abstract

Generic event boundary detection (GEBD) is an important yet challenging task in video understanding, which aims at detecting the moments where humans naturally perceive event boundaries. In this paper, we present a local context modeling and global boundary decoding approach for GEBD task. Local context modeling sub-network is proposed to perceive diverse patterns of generic event boundaries, and it generates powerful video representations and reliable boundary confidence. Based on them, global boundary decoding sub-network is exploited to decode event boundaries from a global view. Our proposed method achieves 85.13% F1-score on Kinetics-GEBD testing set, which achieves a more than 22% F1-score boost compared to the baseline method. The code is available at https://github.com/JackyTown/GEBD_Challenge_CVPR2022.