Bi-directional Training for Composed Image Retrieval via Text Prompt Learning
Open paper
Benchmark
Model
Rank
Results
image-retrieval-on-fashion-iq
BLIP4CIR+Bi
#6
(Recall@10+Recall@50)/2: 55.4
image-retrieval-on-cirr
BLIP4CIR+Bi
#13
(Recall@5+Recall_subset@1)/2: 72.59
Recall@10: 83.88
Rank counts only results with a code link.