Prototype-Based Semantic Consistency Alignment for Domain Adaptive Retrieval

The article introduces Prototype-Based Semantic Consistency Alignment (PSCA), a two-stage framework designed for domain adaptive retrieval that addresses limitations in existing methods. PSCA enhances class-level semantic alignment by utilizing orthogonal prototypes to maximize inter-class separability and improve the reliability of pseudo-labels through geometric proximity. This approach allows for better feature reconstruction and quantization, resulting in improved hash coding quality and superior performance across multiple datasets, which is crucial for practitioners aiming to enhance retrieval systems in domain-shift scenarios.