Revisiting Relevance Feedback for CLIP-based Interactive Image Retrieval
This work addresses the need for personalized image retrieval systems that can handle diverse user preferences without requiring extensive training data, though it is incremental in combining existing techniques.
The authors tackled the problem of adapting image retrieval to individual user preferences by integrating CLIP with classic relevance feedback, achieving competitive accuracy with state-of-the-art metric learning without dataset-specific training.
Many image retrieval studies use metric learning to train an image encoder. However, metric learning cannot handle differences in users' preferences, and requires data to train an image encoder. To overcome these limitations, we revisit relevance feedback, a classic technique for interactive retrieval systems, and propose an interactive CLIP-based image retrieval system with relevance feedback. Our retrieval system first executes the retrieval, collects each user's unique preferences through binary feedback, and returns images the user prefers. Even when users have various preferences, our retrieval system learns each user's preference through the feedback and adapts to the preference. Moreover, our retrieval system leverages CLIP's zero-shot transferability and achieves high accuracy without training. We empirically show that our retrieval system competes well with state-of-the-art metric learning in category-based image retrieval, despite not training image encoders specifically for each dataset. Furthermore, we set up two additional experimental settings where users have various preferences: one-label-based image retrieval and conditioned image retrieval. In both cases, our retrieval system effectively adapts to each user's preferences, resulting in improved accuracy compared to image retrieval without feedback. Overall, our work highlights the potential benefits of integrating CLIP with classic relevance feedback techniques to enhance image retrieval.