# 3D Semantic Segmentation on ScanNet++ 20 classes v2 (test)

Source: [3D Semantic Segmentation on ScanNet++ 20 classes v2 (test)](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-scannet-plus-plus-20-classes-v2-test>)

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

Data updated: 2026-03-05T09:37:21.362Z

Evaluation of 3D Semantic Segmentation performance on the test split of the ScanNet++ 20 classes v2 dataset, typically reporting mIoU and mACC.

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) | ScanNet ➜ ScanNet++ 20 classes v2 (test) |
| Metric | mIoU |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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/3d-semantic-segmentation-on-scannet-plus-plus-20-classes-v2-test.md?metric=mIoU>): Higher is better; unit: PERCENT.

- [mACC](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-scannet-plus-plus-20-classes-v2-test.md?metric=mACC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mIoU | Result date | Report paper |
| --- | --- | --- | --- |
| EmbodiedSplat-fast | 51.66 | Not recorded | 69a90e022d603e6de3e40796 |
| EmbodiedSplat | 44.14 | Not recorded | 69a90e022d603e6de3e40796 |
| Dr. Splat | 39.85 | Not recorded | 69a90e022d603e6de3e40796 |
| EmbodiedSplat-fast | 34.6 | Not recorded | 69a90e022d603e6de3e40796 |
| Occam’s LGS | 34.08 | Not recorded | 69a90e022d603e6de3e40796 |
| EmbodiedSplat | 30.65 | Not recorded | 69a90e022d603e6de3e40796 |
| InstanceGaussian | 29.98 | Not recorded | 69a90e022d603e6de3e40796 |
| OpenGaussian | 25.65 | Not recorded | 69a90e022d603e6de3e40796 |

### EmbodiedSplat-fast

mIoU: 51.66. Result ID: 69a94eb96dbb7075b953209e.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Implementation](<https://github.com/0nandon/EmbodiedSplat>)

Recorded evaluation configuration: {"Input Modality":"RGB-D","Model Speed Variant":"Fast"}. This record does not assert independent reproduction.

### EmbodiedSplat

mIoU: 44.14. Result ID: 69a94eb96dbb7075b953209d.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Implementation](<https://github.com/0nandon/EmbodiedSplat>)

Recorded evaluation configuration: {"Input Modality":"RGB-D","Model Speed Variant":"Standard"}. This record does not assert independent reproduction.

### Dr. Splat

mIoU: 39.85. Result ID: 69a94eb96dbb7075b9532099.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77241e209131cea715d9.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### EmbodiedSplat-fast

mIoU: 34.6. Result ID: 69a94eb96dbb7075b953209c.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Implementation](<https://github.com/0nandon/EmbodiedSplat>)

Recorded evaluation configuration: {"Input Modality":"RGB","Model Speed Variant":"Fast"}. This record does not assert independent reproduction.

### Occam’s LGS

mIoU: 34.08. Result ID: 69a94eb96dbb7075b9532098.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa775d1e209131cea72ba2.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### EmbodiedSplat

mIoU: 30.65. Result ID: 69a94eb96dbb7075b953209b.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Implementation](<https://github.com/0nandon/EmbodiedSplat>)

Recorded evaluation configuration: {"Input Modality":"RGB","Model Speed Variant":"Standard"}. This record does not assert independent reproduction.

### InstanceGaussian

mIoU: 29.98. Result ID: 69a94eb96dbb7075b953209a.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78121e209131cea76a07.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### OpenGaussian

mIoU: 25.65. Result ID: 69a94eb96dbb7075b9532097.

[Reporting paper](<https://www.sota2.com/research/paper/69a90e022d603e6de3e40796.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b383.md>)

[Implementation](<https://3d-aigc.github.io/OpenGaussian>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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