# 3D Indoor Scene Synthesis from Natural Language on Bedroom

Source: [3D Indoor Scene Synthesis from Natural Language on Bedroom](<https://www.sota2.com/research/sota/3d-indoor-scene-synthesis-from-natural-language-on-bedroom>)

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

Data updated: 2026-02-26T06:09:18.460Z

Evaluation of models synthesizing 3D indoor scenes based on natural language prompts, using the Bedroom dataset.

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 indoor scene synthesis from natural language |
| Dataset (including any stated split) | Bedroom |
| Metric | iRecall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [iRecall](<https://www.sota2.com/research/sota/3d-indoor-scene-synthesis-from-natural-language-on-bedroom.md?metric=iRecall>): Higher is better; unit: PERCENT.

- [FID](<https://www.sota2.com/research/sota/3d-indoor-scene-synthesis-from-natural-language-on-bedroom.md?metric=FID>): Lower is better; unit: SCALAR.

- [FID\_CLIP](<https://www.sota2.com/research/sota/3d-indoor-scene-synthesis-from-natural-language-on-bedroom.md?metric=FID_CLIP>): Lower is better; unit: SCALAR.

- [KID](<https://www.sota2.com/research/sota/3d-indoor-scene-synthesis-from-natural-language-on-bedroom.md?metric=KID>): Lower is better; unit: SCALAR.

- [V\_sum\_∩](<https://www.sota2.com/research/sota/3d-indoor-scene-synthesis-from-natural-language-on-bedroom.md?metric=V_sum_%E2%88%A9>): Higher is better; unit: SCALAR.

## Recorded results

| Method | iRecall | Result date | Report paper |
| --- | --- | --- | --- |
| SceneNAT | 70.45 | Not recorded | 6965d1b26343fa78c832393a |
| InstructScene | 66.72 | Not recorded | 6965d1b26343fa78c832393a |
| DiffuScene | 45.98 | Not recorded | 6965d1b26343fa78c832393a |
| ATISS | 31.3 | Not recorded | 6965d1b26343fa78c832393a |

### SceneNAT

iRecall: 70.45. Result ID: 6997cbb25698665ec422dab1.

[Reporting paper](<https://www.sota2.com/research/paper/6965d1b26343fa78c832393a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6965d1b26343fa78c832393a.md>)

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

### InstructScene

iRecall: 66.72. Result ID: 6997cbb25698665ec422dab0.

[Reporting paper](<https://www.sota2.com/research/paper/6965d1b26343fa78c832393a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa721f1e209131cea549db.md>)

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

### DiffuScene

iRecall: 45.98. Result ID: 6997cbb25698665ec422daaf.

[Reporting paper](<https://www.sota2.com/research/paper/6965d1b26343fa78c832393a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa791d1e209131cea7c2fb.md>)

[Implementation](<https://cangmushui.github.io/FreeScene-io/>)

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

### ATISS

iRecall: 31.3. Result ID: 6997cbb25698665ec422daae.

[Reporting paper](<https://www.sota2.com/research/paper/6965d1b26343fa78c832393a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa791d1e209131cea7c2fb.md>)

[Implementation](<https://cangmushui.github.io/FreeScene-io/>)

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.