# 2D Semantic Segmentation on KITTI-360 (val)

Source: [2D Semantic Segmentation on KITTI-360 (val)](<https://www.sota2.com/research/sota/2d-semantic-segmentation-on-kitti-360-val>)

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

Data updated: 2026-02-26T06:10:41.940Z

Evaluation of 2D Semantic Segmentation performance on the validation split of the KITTI-360 dataset, typically measured using mIoU and Pixel Accuracy.

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 | 2D Semantic Segmentation |
| Dataset (including any stated split) | KITTI-360 (val) |
| Metric | mIoU |
| 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](<https://www.sota2.com/research/sota/2d-semantic-segmentation-on-kitti-360-val.md?metric=mIoU>): Higher is better; unit: PERCENT.

- [Pixel Accuracy](<https://www.sota2.com/research/sota/2d-semantic-segmentation-on-kitti-360-val.md?metric=Pixel%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mIoU | Result date | Report paper |
| --- | --- | --- | --- |
| DeepLabV3+ | 68.1 | Not recorded | 694cc50c53fdd5c36173a906 |
| Joint Optics-Sensor-Model Co-Design | 67.21 | Not recorded | 694cc50c53fdd5c36173a906 |
| DeepLabV3 | 66.4 | Not recorded | 694cc50c53fdd5c36173a906 |
| PSPNet | 64.9 | Not recorded | 694cc50c53fdd5c36173a906 |
| FCN-8s | 54 | Not recorded | 694cc50c53fdd5c36173a906 |

### DeepLabV3+

mIoU: 68.1. Result ID: 699813df5698665ec46da492.

[Reporting paper](<https://www.sota2.com/research/paper/694cc50c53fdd5c36173a906.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64b91e209131cea06121.md>)

[Implementation](<https://github.com/tensorflow/models/tree/master/research/deeplab>)

Recorded evaluation configuration: {"Backbone":"Xception-65","Input":"RGB (fixed imaging)","Params (M)":"~42"}. This record does not assert independent reproduction.

### Joint Optics-Sensor-Model Co-Design

mIoU: 67.21. Result ID: 699813df5698665ec46da493.

[Reporting paper](<https://www.sota2.com/research/paper/694cc50c53fdd5c36173a906.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694cc50c53fdd5c36173a906.md>)

Recorded evaluation configuration: {"Input":"RAW-to-task (optics + sensor + model)","Params (M)":"0.47","Epochs":50}. This record does not assert independent reproduction.

### DeepLabV3

mIoU: 66.4. Result ID: 699813df5698665ec46da491.

[Reporting paper](<https://www.sota2.com/research/paper/694cc50c53fdd5c36173a906.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63d91e209131cea014b8.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-101","Input":"RGB (fixed imaging)","Params (M)":"~60"}. This record does not assert independent reproduction.

### PSPNet

mIoU: 64.9. Result ID: 699813df5698665ec46da490.

[Reporting paper](<https://www.sota2.com/research/paper/694cc50c53fdd5c36173a906.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634b1e209131ce9fdc7e.md>)

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

Recorded evaluation configuration: {"Backbone":"ResNet-101","Input":"RGB (fixed imaging)","Params (M)":"~65"}. This record does not assert independent reproduction.

### FCN-8s

mIoU: 54. Result ID: 699813df5698665ec46da48f.

[Reporting paper](<https://www.sota2.com/research/paper/694cc50c53fdd5c36173a906.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa627b1e209131ce9f9766.md>)

[Implementation](<http://fcn.berkeleyvision.org/>)

Recorded evaluation configuration: {"Backbone":"VGG-16","Input":"RGB (fixed imaging)","Params (M)":"~134"}. This record does not assert independent reproduction.

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