MIR-Flickr

The MIR-Flickr dataset is a benchmark that contains images and their associated textual descriptions, used to evaluate cross-modal retrieval methods by assessing the ability to find semantically similar content across different modalities.

Papers using MIR-Flickr (4)

Creating Something From Nothing: Unsupervised Knowledge Distillation For Cross-modal Hashing2020 · 131 citesLearning Discriminative Hashing Codes for Cross-Modal Retrieval based on Multi-view Features2018Creating Something from Nothing: Unsupervised Knowledge Distillation for Cross-Modal Hashing2020Efficient Cross-Modal Retrieval via Deep Binary Hashing and Quantization2022