THU-Warwick Submission for EPIC-KITCHEN Challenge 2025: Semi-Supervised Video Object Segmentation

TL;DR

This paper presents a semi-supervised egocentric video object segmentation method that combines large-scale visual pretraining with depth cues, achieving high accuracy on the VISOR benchmark.

Contribution

It introduces a novel framework integrating SAM2 pretraining and depth cues for improved long-term segmentation in egocentric videos.

Findings

J&F score of 90.1% on VISOR test set

Effective handling of complex scenes and long-term tracking

Integration of visual pretraining with depth cues

Abstract

In this report, we describe our approach to egocentric video object segmentation. Our method combines large-scale visual pretraining from SAM2 with depth-based geometric cues to handle complex scenes and long-term tracking. By integrating these signals in a unified framework, we achieve strong segmentation performance. On the VISOR test set, our method reaches a J&F score of 90.1%.