mask AP: 63.9AP50: 86.7AP75: 71.5AR10: 69.5
Papers with Code · CC BY-SA 4.0
Video Instance Segmentation
The goal of video instance segmentation is simultaneous detection, segmentation and tracking of instances in videos. In words, it is the first time that the image instance segmentation problem is extended to the video domain.
To facilitate research on this new task, a large-scale benchmark called YouTube-VIS, which consists of 2,883 high-resolution YouTube videos, a 40-category label set and 131k high-quality instance masks is built.
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.
Dataset기록된 모델 예시MetricPaperCode
데이터셋
8개논문
mask AP: 60.1AP50: 80.9AP75: 66.5AR10: 64.7
mask AP: 65.3AP50: 87.3AP75: 73.2AR10: 70.3
mask AP: 64.5AP50: 86.1AP75: 72.2AR10: 70.7
mask AP: 60.1AP50: 83.0AP75: 68.4AR10: 65.7
mask AP: 60.3AP50: 81.3AP75: 67.1AR10: 64.5
AP: 55.3
mask AP: 59.8AP50: 82.0AP75: 66.5AR10: 64.4
mask AP: 61.4AP50: 84.1AP75: 68.5AR10: 65.2
mask AP: 60.2AP50: 81.4AP75: 67.6AR10: 64.8
mask AP: 58.4AP50: 79.4AP75: 64.3AR10: 63.6
라이브러리
Papers with Code 리더보드 row에 연결된 구현 저장소입니다.