ELSA-Swin-S

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Medical2 results

Semantic SegmentationonADE20K val
mIoU· 2021-12-23
50.3
best: 62.8 (BEiT-3)
ELSA: Enhanced Local Self-Attention for Vision Transformer arXiv:2112.12786
Semantic SegmentationonADE20K
Validation mIoU· 2021-12-23
50.3
best: 63.6 (ViT-P (InternImage-H))
ELSA: Enhanced Local Self-Attention for Vision Transformer arXiv:2112.12786

Audio2 results

10-shot image generationonADE20K val
mIoU· 2021-12-23
50.3
best: 62.8 (BEiT-3)
ELSA: Enhanced Local Self-Attention for Vision Transformer arXiv:2112.12786
10-shot image generationonADE20K
Validation mIoU· 2021-12-23
50.3
best: 63.6 (ViT-P (InternImage-H))
ELSA: Enhanced Local Self-Attention for Vision Transformer arXiv:2112.12786