# Unsupervised Semantic Segmentation on ImageNet-S50 (test)

Source: [Unsupervised Semantic Segmentation on ImageNet-S50 (test)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:10:17.322Z

Evaluation of Unsupervised Semantic Segmentation performance on the test split of the ImageNet-S50 dataset, reporting metrics including various IoU scores and accuracy measures.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Unsupervised Semantic Segmentation |
| Dataset (including any stated split) | ImageNet-S50 (test) |
| Metric | mIoU |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [mIoU](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=mIoU>): Higher is better; unit: PERCENT.

- [b-mIoU](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=b-mIoU>): Higher is better; unit: PERCENT.

- [Img-Acc](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=Img-Acc>): Higher is better; unit: PERCENT.

- [Fβ](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=F%CE%B2>): Higher is better; unit: PERCENT.

- [mIoU (S.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=mIoU%20(S.)>): Higher is better; unit: PERCENT.

- [mIoU (M.S.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=mIoU%20(M.S.)>): Higher is better; unit: PERCENT.

- [mIoU (M.L.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=mIoU%20(M.L.)>): Higher is better; unit: PERCENT.

- [mIoU (L.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=mIoU%20(L.)>): Higher is better; unit: PERCENT.

- [b-mIoU (S.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=b-mIoU%20(S.)>): Higher is better; unit: PERCENT.

- [b-mIoU (M.S.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=b-mIoU%20(M.S.)>): Higher is better; unit: PERCENT.

- [b-mIoU (M.L.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=b-mIoU%20(M.L.)>): Higher is better; unit: PERCENT.

- [b-mIoU (L.)](<https://www.sota2.com/research/sota/unsupervised-semantic-segmentation-on-imagenet-s50-test.md?metric=b-mIoU%20(L.)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mIoU | Result date | Report paper |
| --- | --- | --- | --- |
| PASS\_p + RC | 41.5 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| PASS\_p + Sal | 41.5 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| PASS\_p | 31.5 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| PASS\_s | 28.6 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| MaskCon | 23.1 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| PICIE | 17.9 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| MDC | 14.6 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| MaskCon+ | 9.2 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| PICIE | 4.6 | Not recorded | 68fa6ddc1e209131cea3c7d5 |
| MDC | 3.7 | Not recorded | 68fa6ddc1e209131cea3c7d5 |

### PASS\_p + RC

mIoU: 41.5. Result ID: 698fe0bb3a6c314849fc7fe2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Implementation](<https://github.com/LUSSeg>)

Recorded evaluation configuration: {"Prior":"S"}. This record does not assert independent reproduction.

### PASS\_p + Sal

mIoU: 41.5. Result ID: 698fe0bb3a6c314849fc7fe3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Implementation](<https://github.com/LUSSeg>)

Recorded evaluation configuration: {"Prior":"S"}. This record does not assert independent reproduction.

### PASS\_p

mIoU: 31.5. Result ID: 698fe0bb3a6c314849fc7fe1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Implementation](<https://github.com/LUSSeg>)

Recorded evaluation configuration: {"Prior":"None"}. This record does not assert independent reproduction.

### PASS\_s

mIoU: 28.6. Result ID: 698fe0bb3a6c314849fc7fe0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Implementation](<https://github.com/LUSSeg>)

Recorded evaluation configuration: {"Prior":"None"}. This record does not assert independent reproduction.

### MaskCon

mIoU: 23.1. Result ID: 698fe0bb3a6c314849fc7fde.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a7b1e209131cea287d5.md>)

[Implementation](<https://github.com/wvangansbeke/Unsupervised-Semantic-Segmentation.git>)

Recorded evaluation configuration: {"Prior":"S"}. This record does not assert independent reproduction.

### PICIE

mIoU: 17.9. Result ID: 698fe0bb3a6c314849fc7fdd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69ac1e209131cea233e6.md>)

Recorded evaluation configuration: {"Prior":"I"}. This record does not assert independent reproduction.

### MDC

mIoU: 14.6. Result ID: 698fe0bb3a6c314849fc7fdb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65861e209131cea0ad87.md>)

Recorded evaluation configuration: {"Prior":"I"}. This record does not assert independent reproduction.

### MaskCon+

mIoU: 9.2. Result ID: 698fe0bb3a6c314849fc7fdf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

Recorded evaluation configuration: {"Prior":"S"}. This record does not assert independent reproduction.

### PICIE

mIoU: 4.6. Result ID: 698fe0bb3a6c314849fc7fdc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69ac1e209131cea233e6.md>)

Recorded evaluation configuration: {"Prior":"None"}. This record does not assert independent reproduction.

### MDC

mIoU: 3.7. Result ID: 698fe0bb3a6c314849fc7fda.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ddc1e209131cea3c7d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65861e209131cea0ad87.md>)

Recorded evaluation configuration: {"Prior":"None"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.