Efficient Vision-Language Pre-training by Cluster Masking

arXiv:2405.08815v1 Announce Type: new
Abstract: We propose a simple strategy for masking image patches during visual-language contrastive learning that improves the quality of the learned representations and the training speed. During each iteration of training, we randomly mask clusters of visually similar image patches, as measured by their raw pixel intensities. This provides an extra learning signal, beyond the contrastive training itself, since it forces a model to predict words for masked visual structures solely from context. It also speeds …

abstract arxiv cluster cs.cv image iteration language masking pixel pre-training quality raw simple speed strategy training type vision vision-language visual