OneFormer (DiNAT-L, single-scale, 1280x1280, COCO-pretrain)

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 SegmentationonADE20K val
AP· uses extra data· 2022-11-10
40.2
best: 44.2 (OneFormer (InternImage-H, emb_dim=1024, single-scale, 896x896, COCO-Pretrained))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Instance SegmentationonADE20K val
APL· uses extra data· 2022-11-10
59.7
best: 64.3 (OneFormer (InternImage-H, emb_dim=1024, single-scale, 896x896, COCO-Pretrained))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Instance SegmentationonADE20K val
APM· uses extra data· 2022-11-10
44.4
best: 49.9 (OneFormer (InternImage-H, emb_dim=1024, single-scale, 896x896, COCO-Pretrained))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Instance SegmentationonADE20K val
APS· uses extra data· 2022-11-10
19.2
best: 23.7 (OneFormer (InternImage-H, emb_dim=1024, single-scale, 896x896, COCO-Pretrained))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220