{"ID":367086,"CreatedAt":"2026-03-04T20:58:37Z","UpdatedAt":"2026-03-04T20:58:37Z","DeletedAt":null,"paper_url":"https://paperswithcode.com/paper/large-scale-unsupervised-semantic","arxiv_id":"2106.03149","title":"Large-scale Unsupervised Semantic Segmentation","abstract":"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.","short_abstract":"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.","url_abs":"https://arxiv.org/abs/2106.03149v3","url_pdf":"https://arxiv.org/pdf/2106.03149v3.pdf","authors":"[\"ShangHua Gao\", \"Zhong-Yu Li\", \"Ming-Hsuan Yang\", \"Ming-Ming Cheng\", \"Junwei Han\", \"Philip Torr\"]","published":"2021-06-06T00:00:00Z","tasks":"[\"Diversity\", \"Representation Learning\", \"Segmentation\", \"Semantic Segmentation\", \"Unsupervised Semantic Segmentation\"]","methods":"[]","has_code":false,"code_links":[{"ID":308153,"CreatedAt":"2026-03-04T21:00:12Z","UpdatedAt":"2026-03-04T21:00:12Z","DeletedAt":null,"paper_id":367086,"paper_url":"https://paperswithcode.com/paper/large-scale-unsupervised-semantic","paper_title":"Large-scale Unsupervised Semantic Segmentation","repo_url":"https://github.com/LUSSeg/ImageNet-S","is_official":true,"mentioned_in_paper":false,"mentioned_in_github":true,"framework":"pytorch","github_stars":0},{"ID":333082,"CreatedAt":"2026-03-04T21:00:12Z","UpdatedAt":"2026-03-04T21:00:12Z","DeletedAt":null,"paper_id":367086,"paper_url":"https://paperswithcode.com/paper/large-scale-unsupervised-semantic","paper_title":"Large-scale Unsupervised Semantic Segmentation","repo_url":"https://github.com/LUSSeg/ImageNetSegModel","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"framework":"pytorch","github_stars":0},{"ID":526954,"CreatedAt":"2026-03-04T21:00:12Z","UpdatedAt":"2026-03-04T21:00:12Z","DeletedAt":null,"paper_id":367086,"paper_url":"https://paperswithcode.com/paper/large-scale-unsupervised-semantic","paper_title":"Large-scale Unsupervised Semantic Segmentation","repo_url":"https://github.com/LUSSeg/PASS","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"framework":"pytorch","github_stars":0}]}