METOR: A Unified Framework for Mutual Enhancement of Objects and Relationships in Open-vocabulary Video Visual Relationship Detection

TL;DR

METOR introduces a unified, query-based framework that jointly enhances object detection and relationship classification in open-vocabulary video analysis, leveraging CLIP and iterative refinement to improve recognition of novel categories.

Contribution

The paper proposes a novel mutual enhancement framework for open-vocabulary video relationship detection, integrating a CLIP-based encoding and iterative refinement to improve accuracy and generalization.

Findings

Achieves state-of-the-art results on VidVRD and VidOR datasets.

Effectively models interdependence between objects and relationships.

Enhances recognition of novel categories in open-vocabulary scenarios.

Abstract

Open-vocabulary video visual relationship detection aims to detect objects and their relationships in videos without being restricted by predefined object or relationship categories. Existing methods leverage the rich semantic knowledge of pre-trained vision-language models such as CLIP to identify novel categories. They typically adopt a cascaded pipeline to first detect objects and then classify relationships based on the detected objects, which may lead to error propagation and thus suboptimal performance. In this paper, we propose Mutual EnhancemenT of Objects and Relationships (METOR), a query-based unified framework to jointly model and mutually enhance object detection and relationship classification in open-vocabulary scenarios. Under this framework, we first design a CLIP-based contextual refinement encoding module that extracts visual contexts of objects and relationships to refine the encoding of text features and object queries, thus improving the generalization of encoding to novel categories. Then we propose an iterative enhancement module to alternatively enhance the representations of objects and relationships by fully exploiting their interdependence to improve recognition performance. Extensive experiments on two public datasets, VidVRD and VidOR, demonstrate that our framework achieves state-of-the-art performance.