Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives

arXiv:2404.11317v1 Announce Type: new
Abstract: The Composed Image Retrieval (CIR) task aims to retrieve target images using a composed query consisting of a reference image and a modified text. Advanced methods often utilize contrastive learning as the optimization objective, which benefits from adequate positive and negative examples. However, the triplet for CIR incurs high manual annotation costs, resulting in limited positive examples. Furthermore, existing methods commonly use in-batch negative sampling, which reduces the negative number available for the model. To …

abstract advanced arxiv benefits cs.ai cs.cv examples however image images improving negative optimization positive query reference retrieval scaling text type via