ViT2Hash: Unsupervised Information-Preserving Hashing
This addresses the problem of information loss in unsupervised hashing for image retrieval, but it is incremental as it builds on existing pre-trained ViT models.
The paper tackles unsupervised image hashing by proposing a method to preserve meaningful information from original data when mapping images to binary codes, achieving significantly higher MAP on three benchmark datasets.
Unsupervised image hashing, which maps images into binary codes without supervision, is a compressor with a high compression rate. Hence, how to preserving meaningful information of the original data is a critical problem. Inspired by the large-scale vision pre-training model, known as ViT, which has shown significant progress for learning visual representations, in this paper, we propose a simple information-preserving compressor to finetune the ViT model for the target unsupervised hashing task. Specifically, from pixels to continuous features, we first propose a feature-preserving module, using the corrupted image as input to reconstruct the original feature from the pre-trained ViT model and the complete image, so that the feature extractor can focus on preserving the meaningful information of original data. Secondly, from continuous features to hash codes, we propose a hashing-preserving module, which aims to keep the semantic information from the pre-trained ViT model by using the proposed Kullback-Leibler divergence loss. Besides, the quantization loss and the similarity loss are added to minimize the quantization error. Our method is very simple and achieves a significantly higher degree of MAP on three benchmark image datasets.