# SVBRDF Multi-view Consistency on 5 synthetic scenes (test)

Source: [SVBRDF Multi-view Consistency on 5 synthetic scenes (test)](<https://www.sota2.com/research/sota/svbrdf-multi-view-consistency-on-5-synthetic-scenes-test>)

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

Data updated: 2026-02-26T05:37:40.690Z

Evaluation of SVBRDF Multi-view Consistency performance on the test split of the 5 synthetic scenes dataset, measured using Basecolor, Roughness, Metallic, and Runtime.

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 | SVBRDF Multi-view Consistency |
| Dataset (including any stated split) | 5 synthetic scenes (test) |
| Metric | Basecolor Consistency |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 10 of 10 |
| 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

- [Basecolor Consistency](<https://www.sota2.com/research/sota/svbrdf-multi-view-consistency-on-5-synthetic-scenes-test.md?metric=Basecolor%20Consistency>): Lower is better; unit: SCALAR.

- [Roughness Consistency](<https://www.sota2.com/research/sota/svbrdf-multi-view-consistency-on-5-synthetic-scenes-test.md?metric=Roughness%20Consistency>): Lower is better; unit: SCALAR.

- [Metallic Consistency](<https://www.sota2.com/research/sota/svbrdf-multi-view-consistency-on-5-synthetic-scenes-test.md?metric=Metallic%20Consistency>): Lower is better; unit: SCALAR.

- [Runtime (s)](<https://www.sota2.com/research/sota/svbrdf-multi-view-consistency-on-5-synthetic-scenes-test.md?metric=Runtime%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | Basecolor Consistency | Result date | Report paper |
| --- | --- | --- | --- |
| UNet-HF | 0.056 | Not recorded | 694238e845614693628503d5 |
| UNet-HF | 0.06 | Not recorded | 694238e845614693628503d5 |
| UNet-RGB | 0.064 | Not recorded | 694238e845614693628503d5 |
| GenPercept | 0.072 | Not recorded | 694238e845614693628503d5 |
| Zeng et al. | 0.073 | Not recorded | 694238e845614693628503d5 |
| GenPercept | 0.075 | Not recorded | 694238e845614693628503d5 |
| UNet-RGB | 0.089 | Not recorded | 694238e845614693628503d5 |
| Kocsis et al. | 0.124 | Not recorded | 694238e845614693628503d5 |
| Zeng et al. | 0.155 | Not recorded | 694238e845614693628503d5 |
| Kocsis et al. | 0.281 | Not recorded | 694238e845614693628503d5 |

### UNet-HF

Basecolor Consistency: 0.056. Result ID: 699843325698665ec49fba54.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

Recorded evaluation configuration: {"Input modalities":"RGB, normals, and depth maps","Model Category":"Regression"}. This record does not assert independent reproduction.

### UNet-HF

Basecolor Consistency: 0.06. Result ID: 699843325698665ec49fba55.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

Recorded evaluation configuration: {"Input modalities":"Training without normals and depth maps","Model Category":"Regression"}. This record does not assert independent reproduction.

### UNet-RGB

Basecolor Consistency: 0.064. Result ID: 699843325698665ec49fba52.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

Recorded evaluation configuration: {"Input modalities":"RGB, normals, and depth maps","Model Category":"Regression"}. This record does not assert independent reproduction.

### GenPercept

Basecolor Consistency: 0.072. Result ID: 699843325698665ec49fba56.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75fd1e209131cea6ac8b.md>)

Recorded evaluation configuration: {"Input modalities":"RGB, normals, and depth maps","Model Category":"Regression"}. This record does not assert independent reproduction.

### Zeng et al.

Basecolor Consistency: 0.073. Result ID: 699843325698665ec49fba51.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

Recorded evaluation configuration: {"Model Category":"Regression","Citation":"\[ZLH\*22\]"}. This record does not assert independent reproduction.

### GenPercept

Basecolor Consistency: 0.075. Result ID: 699843325698665ec49fba57.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75fd1e209131cea6ac8b.md>)

Recorded evaluation configuration: {"Input modalities":"Training without normals and depth maps","Model Category":"Regression"}. This record does not assert independent reproduction.

### UNet-RGB

Basecolor Consistency: 0.089. Result ID: 699843325698665ec49fba53.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

Recorded evaluation configuration: {"Input modalities":"Training without normals and depth maps","Model Category":"Regression"}. This record does not assert independent reproduction.

### Kocsis et al.

Basecolor Consistency: 0.124. Result ID: 699843325698665ec49fba4f.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72a01e209131cea57a5b.md>)

Recorded evaluation configuration: {"Number of predictions averaged (xN)":"10","Model Category":"Generative"}. This record does not assert independent reproduction.

### Zeng et al.

Basecolor Consistency: 0.155. Result ID: 699843325698665ec49fba50.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

Recorded evaluation configuration: {"Model Category":"Generative","Citation":"\[ZDG\*24\]"}. This record does not assert independent reproduction.

### Kocsis et al.

Basecolor Consistency: 0.281. Result ID: 699843325698665ec49fba4e.

[Reporting paper](<https://www.sota2.com/research/paper/694238e845614693628503d5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72a01e209131cea57a5b.md>)

Recorded evaluation configuration: {"Number of predictions averaged (xN)":"1","Model Category":"Generative"}. This record does not assert independent reproduction.

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