# Face Reconstruction on Synthetic Face Dataset

Source: [Face Reconstruction on Synthetic Face Dataset](<https://www.sota2.com/research/sota/face-reconstruction-on-synthetic-face-dataset>)

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

Data updated: 2026-04-14T07:58:10.016Z

Evaluation of Face Reconstruction performance on the Synthetic Face Dataset, measured using Acc, Comp, and NC.

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 | Face Reconstruction |
| Dataset (including any stated split) | Synthetic Face Dataset |
| Metric | Accuracy |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 13 of 13 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/face-reconstruction-on-synthetic-face-dataset.md?metric=Accuracy>): Higher is better; unit: SCALAR.

- [Comp Score](<https://www.sota2.com/research/sota/face-reconstruction-on-synthetic-face-dataset.md?metric=Comp%20Score>): Higher is better; unit: SCALAR.

- [Normalized Correlation (NC)](<https://www.sota2.com/research/sota/face-reconstruction-on-synthetic-face-dataset.md?metric=Normalized%20Correlation%20(NC)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| VGGT | 9.006 | Not recorded | 69ddc9f17d0172140ec7989f |
| PiSSA | 6.89 | Not recorded | 69ddc9f17d0172140ec7989f |
| LoRA | 6.59 | Not recorded | 69ddc9f17d0172140ec7989f |
| PiSSA | 6.526 | Not recorded | 69ddc9f17d0172140ec7989f |
| PiSSA | 6.488 | Not recorded | 69ddc9f17d0172140ec7989f |
| LoRA | 6.393 | Not recorded | 69ddc9f17d0172140ec7989f |
| LoRA | 6.251 | Not recorded | 69ddc9f17d0172140ec7989f |
| Subspace-based LoRA | 5.921 | Not recorded | 69ddc9f17d0172140ec7989f |
| LoRA | 5.767 | Not recorded | 69ddc9f17d0172140ec7989f |
| PiSSA | 5.729 | Not recorded | 69ddc9f17d0172140ec7989f |
| Full | 5.585 | Not recorded | 69ddc9f17d0172140ec7989f |
| Subspace-based LoRA | 4.287 | Not recorded | 69ddc9f17d0172140ec7989f |
| Subspace-based LoRA | 3.831 | Not recorded | 69ddc9f17d0172140ec7989f |

### VGGT

Accuracy: 9.006. Result ID: 69ddf39047f34ce924c52d9a.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa777a1e209131cea736f0.md>)

[Implementation](<https://github.com/facebookresearch/vggt>)

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

### PiSSA

Accuracy: 6.89. Result ID: 69ddf39047f34ce924c52da3.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77dc1e209131cea75a24.md>)

[Implementation](<https://github.com/GraphPKU/PiSSA>)

Recorded evaluation configuration: {"# Trainable Param.":"130.7 M","rank":"128"}. This record does not assert independent reproduction.

### LoRA

Accuracy: 6.59. Result ID: 69ddf39047f34ce924c52d9f.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"# Trainable Param.":"130.7 M","rank":"128"}. This record does not assert independent reproduction.

### PiSSA

Accuracy: 6.526. Result ID: 69ddf39047f34ce924c52da2.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77dc1e209131cea75a24.md>)

[Implementation](<https://github.com/GraphPKU/PiSSA>)

Recorded evaluation configuration: {"# Trainable Param.":"65.3 M","rank":"64"}. This record does not assert independent reproduction.

### PiSSA

Accuracy: 6.488. Result ID: 69ddf39047f34ce924c52da1.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77dc1e209131cea75a24.md>)

[Implementation](<https://github.com/GraphPKU/PiSSA>)

Recorded evaluation configuration: {"# Trainable Param.":"32.7 M","rank":"32"}. This record does not assert independent reproduction.

### LoRA

Accuracy: 6.393. Result ID: 69ddf39047f34ce924c52d9e.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"# Trainable Param.":"65.3 M","rank":"64"}. This record does not assert independent reproduction.

### LoRA

Accuracy: 6.251. Result ID: 69ddf39047f34ce924c52d9d.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"# Trainable Param.":"32.7 M","rank":"32"}. This record does not assert independent reproduction.

### Subspace-based LoRA

Accuracy: 5.921. Result ID: 69ddf39047f34ce924c52da4.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

Recorded evaluation configuration: {"# Trainable Param.":"3.8 M","d":"8"}. This record does not assert independent reproduction.

### LoRA

Accuracy: 5.767. Result ID: 69ddf39047f34ce924c52d9c.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b121e209131cea2bf59.md>)

[Implementation](<https://github.com/microsoft/LoRA>)

Recorded evaluation configuration: {"# Trainable Param.":"16.3 M","rank":"16"}. This record does not assert independent reproduction.

### PiSSA

Accuracy: 5.729. Result ID: 69ddf39047f34ce924c52da0.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77dc1e209131cea75a24.md>)

[Implementation](<https://github.com/GraphPKU/PiSSA>)

Recorded evaluation configuration: {"# Trainable Param.":"16.3 M","rank":"16"}. This record does not assert independent reproduction.

### Full

Accuracy: 5.585. Result ID: 69ddf39047f34ce924c52d9b.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

Recorded evaluation configuration: {"# Trainable Param.":"853.6 M"}. This record does not assert independent reproduction.

### Subspace-based LoRA

Accuracy: 4.287. Result ID: 69ddf39047f34ce924c52da6.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

Recorded evaluation configuration: {"# Trainable Param.":"4.7 M","d":"32"}. This record does not assert independent reproduction.

### Subspace-based LoRA

Accuracy: 3.831. Result ID: 69ddf39047f34ce924c52da5.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddc9f17d0172140ec7989f.md>)

Recorded evaluation configuration: {"# Trainable Param.":"4.0 M","d":"16"}. This record does not assert independent reproduction.

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