# Semantic Segmentation on 20-domain benchmark suite

Source: [Semantic Segmentation on 20-domain benchmark suite](<https://www.sota2.com/research/sota/semantic-segmentation-on-20-domain-benchmark-suite>)

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

Data updated: 2026-02-26T06:09:17.709Z

Evaluation of Semantic Segmentation performance across the 20-domain benchmark suite, reporting metrics such as h-mean.

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 | Semantic Segmentation |
| Dataset (including any stated split) | 20-domain benchmark suite |
| Metric | h-mean |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [h-mean](<https://www.sota2.com/research/sota/semantic-segmentation-on-20-domain-benchmark-suite.md?metric=h-mean>): Higher is better; unit: PERCENT.

## Recorded results

| Method | h-mean | Result date | Report paper |
| --- | --- | --- | --- |
| Oracles | 58.05 | Not recorded | 68fa779d1e209131cea745cf |
| SemLA | 50.57 | Not recorded | 68fa779d1e209131cea745cf |
| Uniform | 48.6 | Not recorded | 68fa779d1e209131cea745cf |
| Zero-shot | 35.56 | Not recorded | 68fa779d1e209131cea745cf |

### Oracles

h-mean: 58.05. Result ID: 699553833badfaffe643a48f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779d1e209131cea745cf.md>)

Recorded evaluation configuration: {"Backbone":"SED \[71\]"}. This record does not assert independent reproduction.

### SemLA

h-mean: 50.57. Result ID: 699553833badfaffe643a491.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779d1e209131cea745cf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779d1e209131cea745cf.md>)

[Implementation](<https://thegoodailab.org/semla>)

Recorded evaluation configuration: {"Backbone":"SED \[71\]","tau":0.05,"K":5}. This record does not assert independent reproduction.

### Uniform

h-mean: 48.6. Result ID: 699553833badfaffe643a490.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779d1e209131cea745cf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74311e209131cea6054b.md>)

Recorded evaluation configuration: {"Backbone":"SED \[71\]"}. This record does not assert independent reproduction.

### Zero-shot

h-mean: 35.56. Result ID: 699553833badfaffe643a48e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa779d1e209131cea745cf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72b11e209131cea57ef4.md>)

[Implementation](<https://ku-cvlab.github.io/CAT-Seg/>)

Recorded evaluation configuration: {"Backbone":"SED \[71\]"}. This record does not assert independent reproduction.

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