In Defense of Clip-based Video Relation Detection

TL;DR

This paper advocates for the clip-based approach in video relation detection, demonstrating that enhanced spatial and temporal context modeling with a hierarchical model outperforms video tubelet methods, achieving state-of-the-art results.

Contribution

The paper introduces a Hierarchical Context Model that significantly improves clip-based VidVRD by better modeling spatial and temporal contexts, surpassing existing video tubelet methods.

Findings

Clip-based methods can outperform video tubelet approaches.

Hierarchical Context Model achieves state-of-the-art results.

Enhanced context modeling improves relation detection accuracy.

Abstract

Video Visual Relation Detection (VidVRD) aims to detect visual relationship triplets in videos using spatial bounding boxes and temporal boundaries. Existing VidVRD methods can be broadly categorized into bottom-up and top-down paradigms, depending on their approach to classifying relations. Bottom-up methods follow a clip-based approach where they classify relations of short clip tubelet pairs and then merge them into long video relations. On the other hand, top-down methods directly classify long video tubelet pairs. While recent video-based methods utilizing video tubelets have shown promising results, we argue that the effective modeling of spatial and temporal context plays a more significant role than the choice between clip tubelets and video tubelets. This motivates us to revisit the clip-based paradigm and explore the key success factors in VidVRD. In this paper, we propose a Hierarchical Context Model (HCM) that enriches the object-based spatial context and relation-based temporal context based on clips. We demonstrate that using clip tubelets can achieve superior performance compared to most video-based methods. Additionally, using clip tubelets offers more flexibility in model designs and helps alleviate the limitations associated with video tubelets, such as the challenging long-term object tracking problem and the loss of temporal information in long-term tubelet feature compression. Extensive experiments conducted on two challenging VidVRD benchmarks validate that our HCM achieves a new state-of-the-art performance, highlighting the effectiveness of incorporating advanced spatial and temporal context modeling within the clip-based paradigm.