Context-I2W: Mapping Images to Context-dependent Words for Accurate Zero-Shot Composed Image Retrieval

Benchmark Model Rank Results
zero-shot-composed-image-retrieval-zs-cir-on-cocoContext-I2W#5Actions Recall@5: 28.5
zero-shot-composed-image-retrieval-zs-cir-on-imagenetContext-I2W#6Average Recall: 20.25
zero-shot-composed-image-retrieval-zs-cir-on-genecisContext-I2W (CLIP L/14)#7A-R@1: 12.7
zero-shot-composed-image-retrieval-zs-cir-on-imagenet-rContext-I2W#9(Recall@10+Recall@50)/2: 20.25
zero-shot-composed-image-retrieval-zs-cir-on-fashion-iqContext-I2W (CLIP L/14)#28(Recall@10+Recall@50)/2: 38.35
zero-shot-composed-image-retrieval-zs-cir-on-circoContext-I2W#33mAP@10: 14.62
zero-shot-composed-image-retrieval-zs-cir-on-cirrContext-I2W (CLIP L/14)#39R@5: 55.1