Seg-B/8 (MS, ViT-B)

Reported on 6 benchmarks across 2 tasks · 1 paper · 2 SOTA

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

Medical3 results

Semantic SegmentationonADE20K val
Pixel Accuracy· 2021-05-12
83.37
best: 83.43 (gSwin-S)
SOTA
Segmenter: Transformer for Semantic Segmentation arXiv:2105.05633
Semantic SegmentationonADE20K val
mIoU· 2021-05-12
49.61
best: 62.8 (BEiT-3)
Segmenter: Transformer for Semantic Segmentation arXiv:2105.05633
Semantic SegmentationonADE20K
Validation mIoU· 2021-05-12
49.61
best: 63.6 (ViT-P (InternImage-H))
Segmenter: Transformer for Semantic Segmentation arXiv:2105.05633

Audio3 results

10-shot image generationonADE20K val
Pixel Accuracy· 2021-05-12
83.37
best: 83.43 (gSwin-S)
SOTA
Segmenter: Transformer for Semantic Segmentation arXiv:2105.05633
10-shot image generationonADE20K val
mIoU· 2021-05-12
49.61
best: 62.8 (BEiT-3)
Segmenter: Transformer for Semantic Segmentation arXiv:2105.05633
10-shot image generationonADE20K
Validation mIoU· 2021-05-12
49.61
best: 63.6 (ViT-P (InternImage-H))
Segmenter: Transformer for Semantic Segmentation arXiv:2105.05633