Semi-Supervised Semantic Segmentation with High- and Low-level Consistency
About
The ability to understand visual information from limited labeled data is an important aspect of machine learning. While image-level classification has been extensively studied in a semi-supervised setting, dense pixel-level classification with limited data has only drawn attention recently. In this work, we propose an approach for semi-supervised semantic segmentation that learns from limited pixel-wise annotated samples while exploiting additional annotation-free images. It uses two network branches that link semi-supervised classification with semi-supervised segmentation including self-training. The dual-branch approach reduces both the low-level and the high-level artifacts typical when training with few labels. The approach attains significant improvement over existing methods, especially when trained with very few labeled samples. On several standard benchmarks - PASCAL VOC 2012, PASCAL-Context, and Cityscapes - the approach achieves new state-of-the-art in semi-supervised learning.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Semantic segmentation | PASCAL VOC 2012 (val) | Mean IoU75.6 | 2040 | |
| Semantic segmentation | Cityscapes (val) | mIoU66 | 572 | |
| Change Detection | LEVIR-CD (test) | -- | 357 | |
| Semantic segmentation | PASCAL VOC (val) | mIoU75.6 | 338 | |
| Semantic segmentation | Cityscapes (val) | mIoU66 | 332 | |
| Change Detection | WHU-CD (test) | IoU76.4 | 286 | |
| Semantic segmentation | Pascal VOC augmented 2012 (val) | mIoU74.5 | 162 | |
| Change Detection | WHU-CD | IoU76.4 | 133 | |
| Semantic segmentation | Cityscapes (val) | mIoU66 | 133 | |
| Semantic segmentation | PASCAL VOC augmented (val) | mIoU73.2 | 122 |