There are two main limitations for existing models: (1) learning of cross-modal representations is restricted to batches; (2) semantically similar samples may be wrongly treated as negative. In this paper, we propose a novel category-level contrastive learning for unsupervised cross-modal hashing,...
networks directly interact with the hash codes to learn the latent discrimination by using instance-level contrast without continuous relaxation, i.e., contrastive hashing learning (𝓛𝒸). The cross-modal ranking loss 𝓛𝑟is utilized to bridge cross-modal hashing learning to cross-modal ...
Unsupervised cross-modal hashing retrievalContrastive learningAdversarial learningCross-modal ranking learningDynamic optimizationCross-modal hashing encodes multimodal data into a common binary space, which can efficiently measure correlations between cross-modal instances. However, most existing cross-modal ...
Although the fact that current methods have some effects, unsupervised cross-modal hashing methods still face several common challenges. First of all, the
另,如果有其他idea很好的跨模态论文,希望你在文章后面留言! Cross-modal Retrieval 一般一个跨模态检索过程可以既包括模态表征,模态转换,模态对齐和联合学习(唯独没有模态融合,基本上不需要融合)。 Adversarial C... Contrastive Adaptation Network for Unsupervised Domain Adaptation...