K-Net (ResNet-101)

Reported on 4 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 Vision4 results

Instance SegmentationonCOCO test-dev
AP50· 2021-06-28
62.8
best: 80.8 (InternImage-H)
K-Net: Towards Unified Image Segmentation arXiv:2106.14855
Instance SegmentationonCOCO test-dev
APL· 2021-06-28
58.8
best: 72.4 (EVA)
K-Net: Towards Unified Image Segmentation arXiv:2106.14855
Instance SegmentationonCOCO test-dev
APM· 2021-06-28
42.7
best: 60.1 (Co-DETR)
K-Net: Towards Unified Image Segmentation arXiv:2106.14855
Instance SegmentationonCOCO test-dev
APS· 2021-06-28
18.7
best: 41.6 (Co-DETR)
K-Net: Towards Unified Image Segmentation arXiv:2106.14855