# Scene-level 3D Reconstruction on Synthetic Room Dataset 10K points with noise

Source: [Scene-level 3D Reconstruction on Synthetic Room Dataset 10K points with noise](<https://www.sota2.com/research/sota/scene-level-3d-reconstruction-on-synthetic-room-dataset-10k-points-with-noise>)

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

Data updated: 2026-02-26T05:39:05.348Z

Evaluation of Scene-level 3D Reconstruction performance using the Synthetic Room Dataset containing 10K noisy points, measured by IoU, Chamfer-L1 distance, Normal Consistency (NC), and F-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 | Scene-level 3D Reconstruction |
| Dataset (including any stated split) | Synthetic Room Dataset 10K points with noise |
| Metric | Chamfer-L1 |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 7 of 7 |
| 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

- [Chamfer-L1](<https://www.sota2.com/research/sota/scene-level-3d-reconstruction-on-synthetic-room-dataset-10k-points-with-noise.md?metric=Chamfer-L1>): Lower is better; unit: SCALAR.

- [IoU](<https://www.sota2.com/research/sota/scene-level-3d-reconstruction-on-synthetic-room-dataset-10k-points-with-noise.md?metric=IoU>): Higher is better; unit: PERCENT.

- [NC](<https://www.sota2.com/research/sota/scene-level-3d-reconstruction-on-synthetic-room-dataset-10k-points-with-noise.md?metric=NC>): Higher is better; unit: PERCENT.

- [F-score](<https://www.sota2.com/research/sota/scene-level-3d-reconstruction-on-synthetic-room-dataset-10k-points-with-noise.md?metric=F-score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Chamfer-L1 | Result date | Report paper |
| --- | --- | --- | --- |
| ALTO | 0.35 | Not recorded | 68fa6e2b1e209131cea3e4d6 |
| POCO | 0.36 | Not recorded | 68fa6e2b1e209131cea3e4d6 |
| ConvONet | 0.42 | Not recorded | 68fa6e2b1e209131cea3e4d6 |
| DP-ConvONet | 0.42 | Not recorded | 68fa6e2b1e209131cea3e4d6 |
| SPSR trimmed | 0.69 | Not recorded | 68fa6e2b1e209131cea3e4d6 |
| ONet | 2.03 | Not recorded | 68fa6e2b1e209131cea3e4d6 |
| SPSR | 2.23 | Not recorded | 68fa6e2b1e209131cea3e4d6 |

### ALTO

Chamfer-L1: 0.35. Result ID: 6990133b3a6c314849fdf20d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

[Implementation](<https://visual.ee.ucla.edu/alto.htm/>)

Recorded evaluation configuration: {"Input points":"10K","Noise":"Added"}. This record does not assert independent reproduction.

### POCO

Chamfer-L1: 0.36. Result ID: 6990133b3a6c314849fdf20c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c181e209131cea321d0.md>)

[Implementation](<https://github.com/valeoai/POCO>)

Recorded evaluation configuration: {"Input points":"10K","Noise":"Added"}. This record does not assert independent reproduction.

### ConvONet

Chamfer-L1: 0.42. Result ID: 6990133b3a6c314849fdf20a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68131e209131cea1a017.md>)

Recorded evaluation configuration: {"Input points":"10K","Noise":"Added"}. This record does not assert independent reproduction.

### DP-ConvONet

Chamfer-L1: 0.42. Result ID: 6990133b3a6c314849fdf20b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68b71e209131cea1dae6.md>)

[Implementation](<https://github.com/dsvilarkovic/dynamic_plane_convolutional_onet>)

Recorded evaluation configuration: {"Input points":"10K","Noise":"Added"}. This record does not assert independent reproduction.

### SPSR trimmed

Chamfer-L1: 0.69. Result ID: 6990133b3a6c314849fdf209.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

Recorded evaluation configuration: {"Input points":"10K","Noise":"Added","Mode":"Trimmed"}. This record does not assert independent reproduction.

### ONet

Chamfer-L1: 2.03. Result ID: 6990133b3a6c314849fdf207.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65a71e209131cea0ba34.md>)

Recorded evaluation configuration: {"Input points":"10K","Noise":"Added"}. This record does not assert independent reproduction.

### SPSR

Chamfer-L1: 2.23. Result ID: 6990133b3a6c314849fdf208.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e2b1e209131cea3e4d6.md>)

Recorded evaluation configuration: {"Input points":"10K","Noise":"Added"}. This record does not assert independent reproduction.

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