U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)

Reported on 4 benchmarks across 2 tasks · 1 paper · 4 SOTA

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

Medical2 results

Semantic SegmentationonPascal VOC 2012 6.25% labeled
Validation mIoU· 2022-03-08
77.21
best: 82 (SemiOVS (w/ PrevMatch, ResNet-101))
SOTA
Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels arXiv:2203.03884
Semantic SegmentationonPASCAL VOC 2012 25% labeled
Validation mIoU· 2022-03-08
79.3
best: 81.3 (SemiOVS (w/ PrevMatch, ResNet-101))
SOTA
Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels arXiv:2203.03884

Audio2 results

10-shot image generationonPascal VOC 2012 6.25% labeled
Validation mIoU· 2022-03-08
77.21
best: 82 (SemiOVS (w/ PrevMatch, ResNet-101))
SOTA
Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels arXiv:2203.03884
10-shot image generationonPASCAL VOC 2012 25% labeled
Validation mIoU· 2022-03-08
79.3
best: 81.3 (SemiOVS (w/ PrevMatch, ResNet-101))
SOTA
Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels arXiv:2203.03884