# Semantic Segmentation on ADE20K 50-50 overlapped (val)

Source: [Semantic Segmentation on ADE20K 50-50 overlapped (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-50-50-overlapped-val>)

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

Data updated: 2026-02-26T06:10:55.538Z

Evaluation of Semantic Segmentation performance on the ADE20K validation set using the specific '50-50 overlapped' class configuration, measured by Mean Intersection over Union (mIoU) across defined class subsets (1-50, 51-100, 101-150) and overall.

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) | ADE20K 50-50 overlapped (val) |
| Metric | mIoU (Classes 1-50) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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 (Classes 1-50)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-50-50-overlapped-val.md?metric=mIoU%20(Classes%201-50)>): Higher is better; unit: PERCENT.

- [mIoU (Classes 51-100)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-50-50-overlapped-val.md?metric=mIoU%20(Classes%2051-100)>): Higher is better; unit: PERCENT.

- [mIoU (Classes 101-150)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-50-50-overlapped-val.md?metric=mIoU%20(Classes%20101-150)>): Higher is better; unit: PERCENT.

- [mIoU (Overall)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-50-50-overlapped-val.md?metric=mIoU%20(Overall)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mIoU (Classes 1-50) | Result date | Report paper |
| --- | --- | --- | --- |
| Joint | 51.1 | Not recorded | 68fa6bfd1e209131cea31762 |
| PLOP | 48.6 | Not recorded | 68fa6bfd1e209131cea31762 |
| RCIL (Ours) | 48.3 | Not recorded | 68fa6bfd1e209131cea31762 |
| MiB | 45.3 | Not recorded | 68fa6bfd1e209131cea31762 |
| ILT | 13.6 | Not recorded | 68fa6bfd1e209131cea31762 |

### Joint

mIoU (Classes 1-50): 51.1. Result ID: 698eb4b6ffc3901de09ff6bc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31762.md>)

Recorded evaluation configuration: {"Continual Learning Scenario":"50-50","Steps":"3 steps"}. This record does not assert independent reproduction.

### PLOP

mIoU (Classes 1-50): 48.6. Result ID: 698eb4b6ffc3901de09ff6ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31762.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa697f1e209131cea22779.md>)

[Implementation](<https://github.com/arthurdouillard/CVPR2021_PLOP>)

Recorded evaluation configuration: {"Continual Learning Scenario":"50-50","Steps":"3 steps"}. This record does not assert independent reproduction.

### RCIL (Ours)

mIoU (Classes 1-50): 48.3. Result ID: 698eb4b6ffc3901de09ff6bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31762.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31762.md>)

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

Recorded evaluation configuration: {"Continual Learning Scenario":"50-50","Steps":"3 steps"}. This record does not assert independent reproduction.

### MiB

mIoU (Classes 1-50): 45.3. Result ID: 698eb4b6ffc3901de09ff6b9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31762.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa673c1e209131cea1516c.md>)

[Implementation](<https://github.com/fcdl94/MiB>)

Recorded evaluation configuration: {"Continual Learning Scenario":"50-50","Steps":"3 steps"}. This record does not assert independent reproduction.

### ILT

mIoU (Classes 1-50): 13.6. Result ID: 698eb4b6ffc3901de09ff6b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31762.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66621e209131cea0ffe7.md>)

[Implementation](<https://lttm.dei.unipd.it/paper_data/IL>)

Recorded evaluation configuration: {"Continual Learning Scenario":"50-50","Steps":"3 steps"}. This record does not assert independent reproduction.

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