Semantic Segmentation on CITYSCAPES 1/4 labeled samples 744 labels (val)
63.87mIoUCutMix
Evaluation results
| Method | Links | |
|---|---|---|
| CutMixBackbone=DeepLab v2, Pre-training=ImageNet, Labeled samples=1/4 (744), Number of runs=52019.06 | 63.87 | |
| CutoutBackbone=DeepLab v2, Pre-training=ImageNet, Labeled samples=1/4 (744), Number of runs=52019.06 | 61.96 | |
| s4GANBackbone=DeepLab v2, Pre-training=ImageNet, Labeled samples=1/4 (744)2019.06 | 61.9 | |
| Baseline (Ours)Backbone=DeepLab v2, Pre-training=ImageNet, Labeled samples=1/4 (744), Number of runs=52019.06 | 60.57 | |
| AdversarialBackbone=DeepLab v2, Pre-training=ImageNet, Labeled samples=1/4 (744)2019.06 | 60.5 | |
| Baseline (Mittal et al.)Backbone=DeepLab v2, Pre-training=ImageNet, Labeled samples=1/4 (744)2019.06 | 60.2 |