# Open Vocabulary 3D Semantic Segmentation on LERF (held-out test views)

Source: [Open Vocabulary 3D Semantic Segmentation on LERF (held-out test views)](<https://www.sota2.com/research/sota/open-vocabulary-3d-semantic-segmentation-on-lerf-held-out-test-views>)

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

Data updated: 2026-04-27T07:20:30.592Z

Evaluation of Open Vocabulary 3D Semantic Segmentation performance on the held-out test views split of the LERF dataset, reporting scene-specific metrics (Ramen, Figurines, Teatime, Waldo Kitchen) and Overall score.

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 | Open Vocabulary 3D Semantic Segmentation |
| Dataset (including any stated split) | LERF (held-out test views) |
| Metric | Ramen 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

- [Ramen mIoU](<https://www.sota2.com/research/sota/open-vocabulary-3d-semantic-segmentation-on-lerf-held-out-test-views.md?metric=Ramen%20mIoU>): Higher is better; unit: PERCENT.

- [Figurines mIoU](<https://www.sota2.com/research/sota/open-vocabulary-3d-semantic-segmentation-on-lerf-held-out-test-views.md?metric=Figurines%20mIoU>): Higher is better; unit: PERCENT.

- [Teatime mIoU](<https://www.sota2.com/research/sota/open-vocabulary-3d-semantic-segmentation-on-lerf-held-out-test-views.md?metric=Teatime%20mIoU>): Higher is better; unit: PERCENT.

- [Waldo Kitchen mIoU](<https://www.sota2.com/research/sota/open-vocabulary-3d-semantic-segmentation-on-lerf-held-out-test-views.md?metric=Waldo%20Kitchen%20mIoU>): Higher is better; unit: PERCENT.

- [Overall mIoU](<https://www.sota2.com/research/sota/open-vocabulary-3d-semantic-segmentation-on-lerf-held-out-test-views.md?metric=Overall%20mIoU>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Ramen mIoU | Result date | Report paper |
| --- | --- | --- | --- |
| NRGS | 52 | Not recorded | 69eeed5e55a3c32af89f95bf |
| OccamLGS | 50.2 | Not recorded | 69eeed5e55a3c32af89f95bf |
| LangSplatV2 | 48.7 | Not recorded | 69eeed5e55a3c32af89f95bf |
| LangSplat | 46.8 | Not recorded | 69eeed5e55a3c32af89f95bf |
| GAGS | 42.1 | Not recorded | 69eeed5e55a3c32af89f95bf |

### NRGS

Ramen mIoU: 52. Result ID: 69ef0e2747f34ce924d4d67f.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed5e55a3c32af89f95bf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eeed5e55a3c32af89f95bf.md>)

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

### OccamLGS

Ramen mIoU: 50.2. Result ID: 69ef0e2747f34ce924d4d67e.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed5e55a3c32af89f95bf.md>)

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

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

### LangSplatV2

Ramen mIoU: 48.7. Result ID: 69ef0e2747f34ce924d4d67d.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed5e55a3c32af89f95bf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1d1e209131cea8c780.md>)

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

### LangSplat

Ramen mIoU: 46.8. Result ID: 69ef0e2747f34ce924d4d67b.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed5e55a3c32af89f95bf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72c11e209131cea58662.md>)

[Implementation](<https://langsplat.github.io/>)

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

### GAGS

Ramen mIoU: 42.1. Result ID: 69ef0e2747f34ce924d4d67c.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed5e55a3c32af89f95bf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa775d1e209131cea72bcb.md>)

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.