Video K-Net (Swin-Base)

Reported on 5 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision5 results

Video Instance SegmentationonYouTube-VIS validation
AP50· 2022-04-10
79
best: 89.3 (CAVIS(ViT-L, Online))
Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation arXiv:2204.04656
Video Instance SegmentationonYouTube-VIS validation
AP75· 2022-04-10
59.6
best: 76.2 (CAVIS(ViT-L, Online))
Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation arXiv:2204.04656
Video Instance SegmentationonYouTube-VIS validation
AR1· 2022-04-10
49.7
best: 58.3 (CAVIS(ViT-L, Online))
Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation arXiv:2204.04656
Video Instance SegmentationonYouTube-VIS validation
AR10· 2022-04-10
59.9
best: 73.7 (DVIS++(ViT-L, Online))
Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation arXiv:2204.04656
Video Instance SegmentationonYouTube-VIS validation
mask AP· 2022-04-10
54.1
best: 68.9 (CAVIS(ViT-L, Online))
Video K-Net: A Simple, Strong, and Unified Baseline for Video Segmentation arXiv:2204.04656