Datasets › ImageNet-S
ImageNet-S (ImageNet Semantic Segmentation)
Powered by the ImageNet dataset, unsupervised learning on large-scale data has made significant advances for classification tasks. There are two major challenges to allowing such an attractive learning modality for segmentation tasks: i) a large-scale benchmark for assessing algorithms is missing; ii) unsupervised shape representation learning is difficult. We propose a new problem of large-scale unsupervised semantic segmentation (LUSS) with a newly created benchmark dataset to track the research progress. Based 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 baseline 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.
Benchmarks archive 2025-07-28
All 6 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.
| First row (archive order) | Paper | Code | ||||
|---|---|---|---|---|---|---|
| Semantic Segmentation | ImageNet-S | TEC (ViT-B/16, 224x224, SSL+FT, mmseg) mIoU (val) 63.2 | Towards Sustainable Self-supervised Learning | sail-sg/tec | 20 | Compare |
| Prompt Engineering | ImageNet-S | POMP Top-1 accuracy % 49.8 | Prompt Pre-Training with Twenty-Thousand Classes for... | amazon-science/prompt-pretraining | 9 | Compare |
| Unsupervised Semantic Segmentation | ImageNet-S-50 | PASS (+Saliency map) mIoU (test) 42.3 | Large-scale Unsupervised Semantic Segmentation | LUSSeg/ImageNet-S +2 | 5 | Compare |
| Unsupervised Semantic Segmentation | ImageNet-S | PASS mIoU (test) 11.0 | Large-scale Unsupervised Semantic Segmentation | LUSSeg/ImageNet-S +2 | 1 | Compare |
| Unsupervised Semantic Segmentation | ImageNet-S-300 | PASS mIoU (test) 18.1 | Large-scale Unsupervised Semantic Segmentation | LUSSeg/ImageNet-S +2 | 1 | Compare |
| Zero-Shot Transfer Image Classification | ImageNet-S | PaLI Accuracy (Private) 63.83 | PaLI: A Jointly-Scaled Multilingual Language-Image Model | google-research/big_vision | 1 | Compare |
Papers archive 2025-07-28
19 shown of 19 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 43. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.
Dataset loaders archive 2025-07-28
3 loaders as listed in the archive; links are outbound and not re-checked here.
Tasks archive 2025-07-28
License archive 2025-07-28
No licence recorded in the archive. Absence here is not a statement about the dataset's terms.
Modalities archive 2025-07-28
Languages archive 2025-07-28
No language tagged.
Variants archive 2025-07-28
- ImageNet-S
- ImageNet-S-300
- ImageNet-S-50
3 variant names, as the archive lists them.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections