# 3D Semantic Segmentation on vKITTI (6-fold val)

Source: [3D Semantic Segmentation on vKITTI (6-fold val)](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val>)

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

Data updated: 2026-02-27T06:37:59.231Z

Evaluation of 3D Semantic Segmentation performance on the 6-fold validation split of the vKITTI dataset, reporting Overall Accuracy (OA) and mean Intersection over Union (mIOU) across various classes.

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 | 3D Semantic Segmentation |
| Dataset (including any stated split) | vKITTI (6-fold val) |
| Metric | OA (Overall Accuracy) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [OA (Overall Accuracy)](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=OA%20(Overall%20Accuracy)>): Higher is better; unit: PERCENT.

- [mIoU (Mean IoU)](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=mIoU%20(Mean%20IoU)>): Higher is better; unit: PERCENT.

- [Terrain IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Terrain%20IoU>): Higher is better; unit: PERCENT.

- [Tree IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Tree%20IoU>): Higher is better; unit: PERCENT.

- [Vegetation IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Vegetation%20IoU>): Higher is better; unit: PERCENT.

- [Building IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Building%20IoU>): Higher is better; unit: PERCENT.

- [Road IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Road%20IoU>): Higher is better; unit: PERCENT.

- [G-Rail IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=G-Rail%20IoU>): Higher is better; unit: PERCENT.

- [T-Sign IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=T-Sign%20IoU>): Higher is better; unit: PERCENT.

- [T-Light IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=T-Light%20IoU>): Higher is better; unit: PERCENT.

- [Pole IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Pole%20IoU>): Higher is better; unit: PERCENT.

- [Misc IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Misc%20IoU>): Higher is better; unit: PERCENT.

- [Truck IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Truck%20IoU>): Higher is better; unit: PERCENT.

- [Car IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Car%20IoU>): Higher is better; unit: PERCENT.

- [Van IoU](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-vkitti-6-fold-val.md?metric=Van%20IoU>): Higher is better; unit: PERCENT.

## Recorded results

| Method | OA (Overall Accuracy) | Result date | Report paper |
| --- | --- | --- | --- |
| MuGNet | 85.1 | Not recorded | 68fa692e1e209131cea209c2 |
| Engelmann et al. | 73.2 | Not recorded | 68fa692e1e209131cea209c2 |
| PointNet | 63.3 | Not recorded | 68fa692e1e209131cea209c2 |

### MuGNet

OA (Overall Accuracy): 85.1. Result ID: 69a13bad35dfcabb00e5a8c8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa692e1e209131cea209c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa692e1e209131cea209c2.md>)

Recorded evaluation configuration: {"Input Features":"XYZ coordinates without RGB values"}. This record does not assert independent reproduction.

### Engelmann et al.

OA (Overall Accuracy): 73.2. Result ID: 69a13bad35dfcabb00e5a8c7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa692e1e209131cea209c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66d31e209131cea12aa9.md>)

Recorded evaluation configuration: {"Input Features":"XYZ coordinates without RGB values"}. This record does not assert independent reproduction.

### PointNet

OA (Overall Accuracy): 63.3. Result ID: 69a13bad35dfcabb00e5a8c6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa692e1e209131cea209c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634f1e209131ce9fe019.md>)

Recorded evaluation configuration: {"Input Features":"XYZ coordinates without RGB values"}. This record does not assert independent reproduction.

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