Large-scale Unsupervised Semantic Segmentation
About
Empowered by large datasets, e.g., ImageNet, unsupervised learning on large-scale data has enabled significant advances for classification tasks. However, whether the large-scale unsupervised semantic segmentation can be achieved remains unknown. There are two major challenges: i) we need a large-scale benchmark for assessing algorithms; ii) we need to develop methods to simultaneously learn category and shape representation in an unsupervised manner. In this work, we propose a new problem of large-scale unsupervised semantic segmentation (LUSS) with a newly created benchmark dataset to help the research progress. Building on the ImageNet dataset, we propose the ImageNet-S dataset with 1.2 million training images and 50k high-quality semantic segmentation annotations for evaluation. Our benchmark has a high data diversity and a clear task objective. We also present a simple yet effective method that works surprisingly well for LUSS. In addition, we benchmark related un/weakly/fully supervised methods accordingly, identifying the challenges and possible directions of LUSS. The benchmark and source code is publicly available at https://github.com/LUSSeg.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Semantic segmentation | PASCAL VOC 2012 (val) | Mean IoU76.1 | 2040 | |
| Object Detection | COCO (val) | mAP40.2 | 613 | |
| Instance Segmentation | COCO (val) | APmk36.5 | 472 | |
| Semantic segmentation | ImageNet S300 (val) | mIoU29.7 | 21 | |
| Semantic segmentation | ImageNet-S (val) | mIoU19.4 | 21 | |
| Semantic segmentation | ImageNet-S (test) | mIoU19.2 | 17 | |
| Image Classification | ImageNet-S 1.0 (test) | -- | 17 | |
| Semantic segmentation | ImageNet-S 1.0 (test) | -- | 14 | |
| Semantic segmentation | ImageNet-S300 (test) | mIoU29.8 | 11 | |
| Unsupervised Semantic Segmentation | ImageNet-S50 (val) | mIoU42.5 | 10 |