Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

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.

Shanghua Gao, Zhong-Yu Li, Ming-Hsuan Yang, Ming-Ming Cheng, Junwei Han, Philip Torr• 2021

Related benchmarks

TaskDatasetResultRank
Semantic segmentationPASCAL VOC 2012 (val)
Mean IoU76.1
2040
Object DetectionCOCO (val)
mAP40.2
613
Instance SegmentationCOCO (val)
APmk36.5
472
Semantic segmentationImageNet S300 (val)
mIoU29.7
21
Semantic segmentationImageNet-S (val)
mIoU19.4
21
Semantic segmentationImageNet-S (test)
mIoU19.2
17
Image ClassificationImageNet-S 1.0 (test)--
17
Semantic segmentationImageNet-S 1.0 (test)--
14
Semantic segmentationImageNet-S300 (test)
mIoU29.8
11
Unsupervised Semantic SegmentationImageNet-S50 (val)
mIoU42.5
10
Showing 10 of 15 rows

Other info

Code

Follow for update