# Semantic Segmentation on Oxford RobotCar cross-city (test)

Source: [Semantic Segmentation on Oxford RobotCar cross-city (test)](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test>)

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

Data updated: 2026-02-26T06:02:20.747Z

Evaluation of Semantic Segmentation performance on the cross-city test split of the Oxford RobotCar dataset, reporting per-class metrics (road, sidewalk, building, etc.) and overall mIoU.

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) | Oxford RobotCar cross-city (test) |
| Metric | Road IoU |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Road IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Road%20IoU>): Higher is better; unit: PERCENT.

- [Sidewalk IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Sidewalk%20IoU>): Higher is better; unit: PERCENT.

- [Building IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Building%20IoU>): Higher is better; unit: PERCENT.

- [Light IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Light%20IoU>): Higher is better; unit: PERCENT.

- [Sign IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Sign%20IoU>): Higher is better; unit: PERCENT.

- [Sky IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Sky%20IoU>): Higher is better; unit: PERCENT.

- [Person IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Person%20IoU>): Higher is better; unit: PERCENT.

- [Automobile IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Automobile%20IoU>): Higher is better; unit: PERCENT.

- [Two-Wheel IoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-oxford-robotcar-cross-city-test.md?metric=Two-Wheel%20IoU>): Higher is better; unit: PERCENT.

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

## Recorded results

| Method | Road IoU | Result date | Report paper |
| --- | --- | --- | --- |
| MRNet | 95.9 | Not recorded | 68fa68801e209131cea1c6ff |
| MRNet+Ours | 95.9 | Not recorded | 68fa68801e209131cea1c6ff |
| AdaptSegNet | 95.1 | Not recorded | 68fa68801e209131cea1c6ff |
| MRNet+Pseudo | 95.1 | Not recorded | 68fa68801e209131cea1c6ff |
| PatchAlign | 94.4 | Not recorded | 68fa68801e209131cea1c6ff |
| Source | 79.2 | Not recorded | 68fa68801e209131cea1c6ff |

### MRNet

Road IoU: 95.9. Result ID: 698d725a8b517eb43321928a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68801e209131cea1c6ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e801e209131cea405f8.md>)

Recorded evaluation configuration: {"adaptation":"baseline"}. This record does not assert independent reproduction.

### MRNet+Ours

Road IoU: 95.9. Result ID: 698d725a8b517eb43321928c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68801e209131cea1c6ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68801e209131cea1c6ff.md>)

Recorded evaluation configuration: {"variance regularization":true,"pseudo-labeling":true}. This record does not assert independent reproduction.

### AdaptSegNet

Road IoU: 95.1. Result ID: 698d725a8b517eb433219288.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68801e209131cea1c6ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67eb1e209131cea19203.md>)

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

Recorded evaluation configuration: {"adaptation":"domain alignment"}. This record does not assert independent reproduction.

### MRNet+Pseudo

Road IoU: 95.1. Result ID: 698d725a8b517eb43321928b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68801e209131cea1c6ff.md>)

Recorded evaluation configuration: {"pseudo-labeling":true}. This record does not assert independent reproduction.

### PatchAlign

Road IoU: 94.4. Result ID: 698d725a8b517eb433219289.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68801e209131cea1c6ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66791e209131cea1067a.md>)

[Implementation](<https://www.nec-labs.com/~mas/adapt-seg/adapt-seg.html>)

Recorded evaluation configuration: {"adaptation":"patch-level alignment"}. This record does not assert independent reproduction.

### Source

Road IoU: 79.2. Result ID: 698d725a8b517eb433219287.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68801e209131cea1c6ff.md>)

Recorded evaluation configuration: {"adaptation":"none"}. This record does not assert independent reproduction.

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