# Semantic Segmentation on Multi-source Domain Generalization GTAV + SYNTHIA (test)

Source: [Semantic Segmentation on Multi-source Domain Generalization GTAV + SYNTHIA (test)](<https://www.sota2.com/research/sota/semantic-segmentation-on-multi-source-domain-generalization-gtav-plus-synthia-test>)

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

Data updated: 2026-02-26T06:10:54.667Z

Evaluation of Semantic Segmentation performance using models trained on a combination of GTAV and SYNTHIA sources, tested on the specified test split, reporting mIoU across target domains (CityScapes, BDD100K, Mapillary).

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) | Multi-source Domain Generalization GTAV + SYNTHIA (test) |
| Metric | mIoU (CityScapes) |
| 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

- [mIoU (CityScapes)](<https://www.sota2.com/research/sota/semantic-segmentation-on-multi-source-domain-generalization-gtav-plus-synthia-test.md?metric=mIoU%20(CityScapes)>): Higher is better; unit: PERCENT.

- [mIoU (BDD100K)](<https://www.sota2.com/research/sota/semantic-segmentation-on-multi-source-domain-generalization-gtav-plus-synthia-test.md?metric=mIoU%20(BDD100K)>): Higher is better; unit: PERCENT.

- [mIoU (Mapillary)](<https://www.sota2.com/research/sota/semantic-segmentation-on-multi-source-domain-generalization-gtav-plus-synthia-test.md?metric=mIoU%20(Mapillary)>): Higher is better; unit: PERCENT.

- [mIoU (Mean)](<https://www.sota2.com/research/sota/semantic-segmentation-on-multi-source-domain-generalization-gtav-plus-synthia-test.md?metric=mIoU%20(Mean)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mIoU (CityScapes) | Result date | Report paper |
| --- | --- | --- | --- |
| SHADE | 47.43 | Not recorded | 68fa6d171e209131cea379c8 |
| ISW | 37.69 | Not recorded | 68fa6d171e209131cea379c8 |
| IBN-Net | 35.55 | Not recorded | 68fa6d171e209131cea379c8 |
| Baseline | 35.46 | Not recorded | 68fa6d171e209131cea379c8 |

### SHADE

mIoU (CityScapes): 47.43. Result ID: 698f7c663a6c314849f9254b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379c8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379c8.md>)

[Implementation](<https://github.com/HeliosZhao/SHADE2>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Sources":"GTAV + SYNTHIA"}. This record does not assert independent reproduction.

### ISW

mIoU (CityScapes): 37.69. Result ID: 698f7c663a6c314849f9254a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379c8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b31e209131cea238c6.md>)

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

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Sources":"GTAV + SYNTHIA"}. This record does not assert independent reproduction.

### IBN-Net

mIoU (CityScapes): 35.55. Result ID: 698f7c663a6c314849f92549.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379c8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa673e1e209131cea152bb.md>)

[Implementation](<https://github.com/XingangPan/IBN-Net>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Sources":"GTAV + SYNTHIA"}. This record does not assert independent reproduction.

### Baseline

mIoU (CityScapes): 35.46. Result ID: 698f7c663a6c314849f92548.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea379c8.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Sources":"GTAV + SYNTHIA"}. This record does not assert independent reproduction.

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