English

Large-scale Unsupervised Semantic Segmentation

Computer Vision and Pattern Recognition 2022-11-04 v3

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.

Keywords

Cite

@article{arxiv.2106.03149,
  title  = {Large-scale Unsupervised Semantic Segmentation},
  author = {Shanghua Gao and Zhong-Yu Li and Ming-Hsuan Yang and Ming-Ming Cheng and Junwei Han and Philip Torr},
  journal= {arXiv preprint arXiv:2106.03149},
  year   = {2022}
}

Comments

Benchmark and Source Code: https://github.com/LUSSeg

R2 v1 2026-06-24T02:53:02.920Z