Semantic Segmentation on PASCAL VOC augmented 2012 (100% labeled)
76.5mIoUCompL (DenseCL)
Evaluation results
| Method | Links | |
|---|---|---|
| CompL (DenseCL)Training Strategy=Joint Training, Auxiliary Task=DenseCL, Backbone=ResNet-1012022.10 | 76.5 | |
| Semseg BaselineTraining Strategy=Supervised Baseline, Backbone=ResNet-1012022.10 | 75.67 | |
| CompL (DenseCL)Training Strategy=Joint Training, Auxiliary Task=DenseCL, Backbone=ResNet-502022.10 | 74.2 | |
| Semseg BaselineTraining Strategy=Supervised Baseline, Backbone=ResNet-502022.10 | 74.16 | |
| Semseg + DenseCLTraining Strategy=Joint Training (CompL), Auxiliary Task=DenseCL, Backbone=ResNet-262022.10 | 71.16 | |
| Semseg + MoCoTraining Strategy=Joint Training (CompL), Auxiliary Task=MoCo, Backbone=ResNet-262022.10 | 71.15 | |
| Semseg + RotTraining Strategy=Joint Training (CompL), Auxiliary Task=Rotation, Backbone=ResNet-262022.10 | 70.52 | |
| SemsegTraining Strategy=Supervised Baseline, Backbone=ResNet-262022.10 | 70.42 | |
| DenseCL → Semseg + DenseCLTraining Strategy=Transfer + Joint Training (CompL), Auxiliary Task=DenseCL, Backbone=ResNet-262022.10 | 69.81 | |
| DenseCL → SemsegTraining Strategy=Transfer Learning, Pre-training Task=DenseCL, Backbone=ResNet-262022.10 | 69.56 | |
| MoCo → SemsegTraining Strategy=Transfer Learning, Pre-training Task=MoCo, Backbone=ResNet-262022.10 | 68.09 | |
| Rot → SemsegTraining Strategy=Transfer Learning, Pre-training Task=Rotation, Backbone=ResNet-262022.10 | 39.46 |