Image-to-text R@1: 78.8Image-to-text R@5: 96.1Image-to-text R@10: 98.6Text-to-image R@1: 63.7
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.
개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.
벤치마크
Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.
데이터셋
3개논문
Image-to-text R@1: 79.6Image-to-text R@5: 96.1Image-to-text R@10: 98.7Text-to-image R@1: 64.7
Image-to-text R@1: 78.9Image-to-text R@5: 96.3Image-to-text R@10: 98.6Text-to-image R@1: 63.3
Image-to-text R@1: 79.6Image-to-text R@5: 96.5Image-to-text R@10: 98.5Text-to-image R@1: 64.4
Image-to-text R@1: 79.8Image-to-text R@5: 96.6Image-to-text R@10: 98.9Text-to-image R@1: 63.8
Image-to-text R@1: 80.2Image-to-text R@5: 96.3Image-to-text R@10: 98.5Text-to-image R@1: 64.2
Image-to-text R@1: 79.5Image-to-text R@5: 96.4Image-to-text R@10: 98.9Text-to-image R@1: 64.4
Image-to-text R@1: 80.3Image-to-text R@5: 96.2Image-to-text R@10: 98.5Text-to-image R@1: 63.2
Image-to-text R@1: 78.1Image-to-text R@5: 97.2Image-to-text R@10: 98.8Text-to-image R@1: 64.3
Image-to-text R@1: 78.5Image-to-text R@5: 96.8Image-to-text R@10: 98.8Text-to-image R@1: 63.8
Image-to-text R@1: 80.9Image-to-text R@5: 96.6Image-to-text R@10: 98.8Text-to-image R@1: 65.2
Image-to-text R@1: 79.5Image-to-text R@5: 97.2Image-to-text R@10: 99.0Text-to-image R@1: 63.7
라이브러리
Papers with Code 리더보드 row에 연결된 구현 저장소입니다.