# 3D LiDAR Segmentation on SemanticKITTI, nuScenes, and SemanticPOSS

Source: [3D LiDAR Segmentation on SemanticKITTI, nuScenes, and SemanticPOSS](<https://www.sota2.com/research/sota/3d-lidar-segmentation-on-semantickitti-nuscenes-and-semanticposs>)

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

Data updated: 2026-03-11T08:07:58.168Z

Evaluation of 3D LiDAR Semantic Segmentation performance across the evaluation sets of SemanticKITTI, nuScenes, and SemanticPOSS, reporting combined mIoU, AM, and HM scores.

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 LiDAR Segmentation |
| Dataset (including any stated split) | SemanticKITTI, nuScenes, and SemanticPOSS (Evaluation Sets) |
| Metric | mIoU (SemanticKITTI) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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 (SemanticKITTI)](<https://www.sota2.com/research/sota/3d-lidar-segmentation-on-semantickitti-nuscenes-and-semanticposs.md?metric=mIoU%20(SemanticKITTI)>): Higher is better; unit: PERCENT.

- [mIoU (nuScenes)](<https://www.sota2.com/research/sota/3d-lidar-segmentation-on-semantickitti-nuscenes-and-semanticposs.md?metric=mIoU%20(nuScenes)>): Higher is better; unit: PERCENT.

- [mIoU (SemanticPOSS)](<https://www.sota2.com/research/sota/3d-lidar-segmentation-on-semantickitti-nuscenes-and-semanticposs.md?metric=mIoU%20(SemanticPOSS)>): Higher is better; unit: PERCENT.

- [AM](<https://www.sota2.com/research/sota/3d-lidar-segmentation-on-semantickitti-nuscenes-and-semanticposs.md?metric=AM>): Higher is better; unit: PERCENT.

- [HM](<https://www.sota2.com/research/sota/3d-lidar-segmentation-on-semantickitti-nuscenes-and-semanticposs.md?metric=HM>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mIoU (SemanticKITTI) | Result date | Report paper |
| --- | --- | --- | --- |
| DuNe | 56.86 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| DuNe | 53.2 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| DuNe | 52.37 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| NPN | 52.05 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| NPN | 48.72 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| NPN | 41.78 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| DISC | 41.42 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| DISC | 36.68 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| DISC | 33.64 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| TCL | 19.59 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| TCL | 14.14 | Not recorded | 68fa7bfe1e209131cea8bc27 |
| TCL | 10.37 | Not recorded | 68fa7bfe1e209131cea8bc27 |

### DuNe

mIoU (SemanticKITTI): 56.86. Result ID: 69b122da6dbb7075b95b62e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

Recorded evaluation configuration: {"Noise Ratio":"10%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### DuNe

mIoU (SemanticKITTI): 53.2. Result ID: 69b122da6dbb7075b95b62ec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

Recorded evaluation configuration: {"Noise Ratio":"20%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### DuNe

mIoU (SemanticKITTI): 52.37. Result ID: 69b122da6dbb7075b95b62f0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

Recorded evaluation configuration: {"Noise Ratio":"50%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### NPN

mIoU (SemanticKITTI): 52.05. Result ID: 69b122da6dbb7075b95b62e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71761e209131cea515fd.md>)

Recorded evaluation configuration: {"Noise Ratio":"10%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### NPN

mIoU (SemanticKITTI): 48.72. Result ID: 69b122da6dbb7075b95b62eb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71761e209131cea515fd.md>)

Recorded evaluation configuration: {"Noise Ratio":"20%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### NPN

mIoU (SemanticKITTI): 41.78. Result ID: 69b122da6dbb7075b95b62ef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71761e209131cea515fd.md>)

Recorded evaluation configuration: {"Noise Ratio":"50%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### DISC

mIoU (SemanticKITTI): 41.42. Result ID: 69b122da6dbb7075b95b62e6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

Recorded evaluation configuration: {"Noise Ratio":"10%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### DISC

mIoU (SemanticKITTI): 36.68. Result ID: 69b122da6dbb7075b95b62ea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

Recorded evaluation configuration: {"Noise Ratio":"20%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### DISC

mIoU (SemanticKITTI): 33.64. Result ID: 69b122da6dbb7075b95b62ee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

Recorded evaluation configuration: {"Noise Ratio":"50%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### TCL

mIoU (SemanticKITTI): 19.59. Result ID: 69b122da6dbb7075b95b62e5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec91e209131cea421ed.md>)

[Implementation](<https://github.com/Hzzone/TCL>)

Recorded evaluation configuration: {"Noise Ratio":"10%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### TCL

mIoU (SemanticKITTI): 14.14. Result ID: 69b122da6dbb7075b95b62e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec91e209131cea421ed.md>)

[Implementation](<https://github.com/Hzzone/TCL>)

Recorded evaluation configuration: {"Noise Ratio":"20%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

### TCL

mIoU (SemanticKITTI): 10.37. Result ID: 69b122da6dbb7075b95b62ed.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bfe1e209131cea8bc27.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec91e209131cea421ed.md>)

[Implementation](<https://github.com/Hzzone/TCL>)

Recorded evaluation configuration: {"Noise Ratio":"50%","Training Dataset":"SemanticKITTI"}. This record does not assert independent reproduction.

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