ViT-Adapter-L (HTC++, BEiTv2, O365, multi-scale)

Reported on 2 benchmarks across 1 task · 1 paper · 1 SOTA

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

Computer Vision2 results

Instance SegmentationonCOCO minival
mask AP· uses extra data· 2022-05-17
54.2
best: 56.6 (Co-DETR)
SOTA
Vision Transformer Adapter for Dense Predictions arXiv:2205.08534
Instance SegmentationonCOCO test-dev
mask AP· uses extra data· 2022-05-17
54.5
best: 57.1 (Co-DETR)
Vision Transformer Adapter for Dense Predictions arXiv:2205.08534