Dissecting Deep Metric Learning Losses for Image-Text Retrieval
Open paper
Benchmark
Model
Rank
Results
cross-modal-retrieval-on-coco-2014
VSE-Gradient
#7
Text-to-image R@1: 63.6
Text-to-image R@5: 86.0
…
cross-modal-retrieval-on-flickr30k
VSE-Gradient
#6
Image-to-text R@1: 97.0
Image-to-text R@5: 99.6
Image-to-text R@10: 100
…
Rank counts only results with a code link.