What's in the Flow? Exploiting Temporal Motion Cues for Unsupervised Generic Event Boundary Detection

TL;DR

This paper introduces FlowGEBD, a non-parametric, unsupervised method that leverages optical flow to detect generic event boundaries in videos, outperforming neural models on key datasets.

Contribution

The paper presents FlowGEBD, a novel unsupervised, non-parametric approach using motion cues for event boundary detection, achieving state-of-the-art results.

Findings

FlowGEBD surpasses neural models on Kinetics-GEBD with F1@0.05 of 0.713.

FlowGEBD achieves an average F1 score of 0.623 on TAPOS.

Motion cues alone are sufficient for high-performance event boundary detection.

Abstract

Generic Event Boundary Detection (GEBD) task aims to recognize generic, taxonomy-free boundaries that segment a video into meaningful events. Current methods typically involve a neural model trained on a large volume of data, demanding substantial computational power and storage space. We explore two pivotal questions pertaining to GEBD: Can non-parametric algorithms outperform unsupervised neural methods? Does motion information alone suffice for high performance? This inquiry drives us to algorithmically harness motion cues for identifying generic event boundaries in videos. In this work, we propose FlowGEBD, a non-parametric, unsupervised technique for GEBD. Our approach entails two algorithms utilizing optical flow: (i) Pixel Tracking and (ii) Flow Normalization. By conducting thorough experimentation on the challenging Kinetics-GEBD and TAPOS datasets, our results establish FlowGEBD as the new state-of-the-art (SOTA) among unsupervised methods. FlowGEBD exceeds the neural models on the Kinetics-GEBD dataset by obtaining an F1@0.05 score of 0.713 with an absolute gain of 31.7% compared to the unsupervised baseline and achieves an average F1 score of 0.623 on the TAPOS validation dataset.