Graph-Based Cross-Domain Knowledge Distillation for Cross-Dataset Text-to-Image Person Retrieval

2025-01-25

AI Summary

This work addresses cross-dataset text-to-image person retrieval under unsupervised domain adaptation, where no labels are available in the target domain. Method: We propose a graph-structured guided cross-domain knowledge distillation framework. It constructs a cross-domain heterogeneous graph to model inter-modal (text–image) and inter-domain relationships, designs a graph-driven multimodal message propagation mechanism, and introduces a contrastive momentum-based knowledge distillation module for end-to-end adaptation without target-domain annotations. Contribution/Results: By innovatively integrating graph neural networks, online knowledge distillation, and momentum-updated feature queues, our method achieves significant improvements over state-of-the-art approaches on three public benchmarks. Extensive experiments demonstrate its robustness and efficiency in unsupervised cross-domain cross-modal retrieval.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Image and Video Retrieval
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Video surveillance systems are crucial components for ensuring public safety and management in smart city. As a fundamental task in video surveillance, text-to-image person retrieval aims to retrieve the target person from an image gallery that best matches the given text description. Most existing text-to-image person retrieval methods are trained in a supervised manner that requires sufficient labeled data in the target domain. However, it is common in practice that only unlabeled data is available in the target domain due to the difficulty and cost of data annotation, which limits the generalization of existing methods in practical application scenarios. To address this issue, we propose a novel unsupervised domain adaptation method, termed Graph-Based Cross-Domain Knowledge Distillation (GCKD), to learn the cross-modal feature representation for text-to-image person retrieval in a cross-dataset scenario. The proposed GCKD method consists of two main components. Firstly, a graph-based multi-modal propagation module is designed to bridge the cross-domain correlation among the visual and textual samples. Secondly, a contrastive momentum knowledge distillation module is proposed to learn the cross-modal feature representation using the online knowledge distillation strategy. By jointly optimizing the two modules, the proposed method is able to achieve efficient performance for cross-dataset text-to-image person retrieval. acExtensive experiments on three publicly available text-to-image person retrieval datasets demonstrate the effectiveness of the proposed GCKD method, which consistently outperforms the state-of-the-art baselines.

Problem

• Video Surveillance
• Text-to-Image Retrieval
• Unlabeled Data

Innovation

• GCKD
• Unsupervised Learning
• Cross-Dataset Person Matching