Text-Phase Synergy Network with Dual Priors for Unsupervised Cross-Domain Image Retrieval

2026-03-13

AI Summary

This work addresses semantic degradation in unsupervised cross-domain image retrieval, which arises from inaccurate pseudo-label semantics and entanglement between domain-specific and semantic information. To mitigate these issues, the authors propose a dual-prior collaboration mechanism that leverages CLIP-generated domain-specific textual prompts to provide precise semantic supervision, while simultaneously introducing domain-invariant phase features to disentangle domain and semantic representations. By synergistically integrating these two components, the method optimizes cross-domain feature representations without requiring labeled data, thereby preserving semantic integrity and aligning feature distributions across domains. Extensive experiments demonstrate that the proposed approach significantly outperforms state-of-the-art methods on multiple unsupervised cross-domain image retrieval benchmarks.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

This paper studies unsupervised cross-domain image retrieval (UCDIR), which aims to retrieve images of the same category across different domains without relying on labeled data. Existing methods typically utilize pseudo-labels, derived from clustering algorithms, as supervisory signals for intra-domain representation learning and cross-domain feature alignment. However, these discrete pseudo-labels often fail to provide accurate and comprehensive semantic guidance. Moreover, the alignment process frequently overlooks the entanglement between domain-specific and semantic information, leading to semantic degradation in the learned representations and ultimately impairing retrieval performance. This paper addresses the limitations by proposing a Text-Phase Synergy Network with Dual Priors(TPSNet). Specifically, we first employ CLIP to generate a set of class-specific prompts per domain, termed as domain prompt, serving as a text prior that offers more precise semantic supervision. In parallel, we further introduce a phase prior, represented by domain-invariant phase features, which is integrated into the original image representations to bridge the domain distribution gaps while preserving semantic integrity. Leveraging the synergy of these dual priors, TPSNet significantly outperforms state-of-the-art methods on UCDIR benchmarks.

Problem

• unsupervised cross-domain image retrieval
• pseudo-labels
• semantic degradation
• domain alignment
• semantic guidance

Innovation

• Text-Phase Synergy
• Dual Priors
• Unsupervised Cross-Domain Image Retrieval
• Domain-Invariant Phase Features
• CLIP-based Semantic Prompting