The objective in this paper is to improve the performance of text-to-image retrieval. To this end, we introduce a new framework that can boost the performance of large-scale pre-trained vision-language models, so that they can be used for text-to-image re-ranking. The approach, Enhanced Language-Image Pre-training (ELIP), uses the text query, via a simple MLP mapping network, to predict a set of visual prompts to condition the ViT image encoding. ELIP can easily be applied to the commonly used CLIP, SigLIP and BLIP-2 networks. To train the architecture with limited computing resources, we develop a 'student friendly' best practice, involving global hard sample mining, and curation of a large-scale dataset. On the evaluation side, we set up two new out-of-distribution (OOD) benchmarks, Occluded COCO and ImageNet-R, to assess the zero-shot generalisation of the models to different domains. The results demonstrate that ELIP significantly boosts CLIP/SigLIP/SigLIP-2 text-to-image retrieval performance and outperforms BLIP-2 on several benchmarks, as well as providing an easy means to adapt to OOD datasets.
@article{arxiv.2502.15682,
title = {ELIP: Enhanced Visual-Language Foundation Models for Image Retrieval},
author = {Guanqi Zhan and Yuanpei Liu and Kai Han and Weidi Xie and Andrew Zisserman},
journal= {arXiv preprint arXiv:2502.15682},
year = {2025}
}
Comments
Accepted by CBMI 2025 (IEEE International Conference on Content-Based Multimedia Indexing)