Deep Rank Cross-Modal Hashing with Semantic Consistent for Image-Text Retrieval

Xiaoqing Liu, Huanqiang Zeng, Yifan Shi, Jianqing Zhu, Kai-Kuang MaICASSP 2022 · OpenReview

Abstract

Cross-modal hashing retrieval approaches maps heterogeneous multi-modal data into a common hamming space to achieve efficient and flexible retrieval performance. However, existing cross-modal methods mainly exploit feature-level similarity between multi-modal data, the label-level similarity and relative ranking relationship between adjacent instances have been ignored. To address these problems, we propose a novel Deep Rank Cross-modal Hashing(DRCH) method that fully explores the intra-modal semantic similarity relationship. Firstly, DRCH preserves semantic similarity by combining both label-level and feature-level information. Secondly, the inherent gap between modalities are narrowed by proposing a ranking alignment loss function. Finally, the compact and efficient hash codes are optimized from the common semantic space. Extensive experiments on two real-world image-text retrieval datasets demonstrate the superiority of DRCH compared with several state-of-the-art(SOTA) methods.
BibTeX
@inproceedings{liu2022deep,
  author = {Xiaoqing Liu and Huanqiang Zeng and Yifan Shi and Jianqing Zhu and Kai-Kuang Ma},
  title = {{Deep Rank Cross-Modal Hashing with Semantic Consistent for Image-Text Retrieval}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2022},
  url = {https://openreview.net/forum?id=YejUzXyCAf},
}

Loading related content...

Figure