OpenCodePapers

image-retrieval-on-muge-retrieval

Image Retrieval
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeMean RecallR@1R@10R@5ModelNameReleaseDate
Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese✓ Link83.668.993.188.7CN-CLIP (ViT-H/14)2022-11-02
Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese✓ Link81.365.392.186.7CN-CLIP (ViT-L/14@336px)2022-11-02
Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese✓ Link80.163.391.385.6CN-CLIP (ViT-L/14)2022-11-02
CCMB: A Large-scale Chinese Cross-modal Benchmark✓ Link77.560.189.482.9R2D2 (ViT-L/14)2022-05-08
Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese✓ Link77.458.490.083.6CN-CLIP (ViT-B/16)2022-11-02
Wukong: A 100 Million Large-scale Chinese Cross-modal Pre-training Benchmark✓ Link72.152.785.677.9Wukong (ViT-L/14)2022-02-14
Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese✓ Link69.248.684.075.1CN-CLIP (RN50)2022-11-02
CCMB: A Large-scale Chinese Cross-modal Benchmark✓ Link68.747.483.575.1R2D2 (ViT-B)2022-05-08
Wukong: A 100 Million Large-scale Chinese Cross-modal Pre-training Benchmark✓ Link61.239.277.466.9Wukong (ViT-B/32)2022-02-14