Image-to-text R@1: 39.5Image-to-text R@5: 64.0Image-to-text R@10: 73.1Text-to-image R@1: 40.6
Papers with Code · CC BY-SA 4.0
Cross-modal retrieval with noisy correspondence
Noisy correspondence learning aims to eliminate the negative impact of the mismatched pairs (e.g., false positives/negatives) instead of annotation errors in several tasks.
논문
Image-to-text R@1: 39.5Image-to-text R@5: 64.5Image-to-text R@10: 73.5Text-to-image R@1: 40.3
Image-to-text R@1: 78.1Image-to-text R@5: 94.4Image-to-text R@10: 97.5Text-to-image R@1: 60.4
Image-to-text R@1: 77.9Image-to-text R@5: 95.4Image-to-text R@10: 98.3Text-to-image R@1: 60.9
Image-to-text R@1: 77.4Image-to-text R@5: 95.0Image-to-text R@10: 97.3Text-to-image R@1: 58.7
Image-to-text R@1: 74.2Image-to-text R@5: 91.8Image-to-text R@10: 96.9Text-to-image R@1: 55.6
Image-to-text R@1: 77.5Image-to-text R@5: 93.8Image-to-text R@10: 97.0Text-to-image R@1: 56.1
Image-to-text R@1: 76.2Image-to-text R@5: 95.8Image-to-text R@10: 98.3Text-to-image R@1: 60.5
Image-to-text R@1: 77.9Image-to-text R@5: 95.2Image-to-text R@10: 97.8Text-to-image R@1: 59.8
Image-to-text R@1: 78.3Image-to-text R@5: 94.6Image-to-text R@10: 97.8Text-to-image R@1: 60.1
Image-to-text R@1: 79.3Image-to-text R@5: 94.6Image-to-text R@10: 97.8Text-to-image R@1: 60.8
Image-to-text R@1: 77.4Image-to-text R@5: 94.9Image-to-text R@10: 97.6Text-to-image R@1: 59.6
Image-to-text R@1: 79.5Image-to-text R@5: 94.2Image-to-text R@10: 97.9Text-to-image R@1: 61.2
Image-to-text R@1: 80.3Image-to-text R@5: 95.3Image-to-text R@10: 97.8Text-to-image R@1: 61.6
Image-to-text R@1: 79.2Image-to-text R@5: 95.0Image-to-text R@10: 96.9Text-to-image R@1: 59.4
Image-to-text R@1: 78.4Image-to-text R@5: 95.8Image-to-text R@10: 97.8Text-to-image R@1: 59.8