# Video Restoration on UDM10

Source: [Video Restoration on UDM10](<https://www.sota2.com/research/sota/video-restoration-on-udm10>)

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

Data updated: 2026-03-25T14:02:12.643Z

Evaluation of video restoration performance on the UDM10 dataset using a comprehensive set of perceptual and learned quality assessment metrics (LPIPS, DISTS, CLIPIQA, MUSIQ, MANIQA, DOVER, Ewarp, VQ-Insight).

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 | Video Restoration |
| Dataset (including any stated split) | UDM10 |
| Metric | LPIPS |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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

- [LPIPS](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=LPIPS>): Lower is better; unit: SCALAR.

- [DISTS](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=DISTS>): Lower is better; unit: SCALAR.

- [CLIPIQA](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=CLIPIQA>): Higher is better; unit: SCALAR.

- [MUSIQ Score](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=MUSIQ%20Score>): Higher is better; unit: SCALAR.

- [MANIQA](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=MANIQA>): Higher is better; unit: SCALAR.

- [DOVER](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=DOVER>): Higher is better; unit: SCALAR.

- [Ewarp Error](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=Ewarp%20Error>): Lower is better; unit: SCALAR.

- [VQ-Insight Score](<https://www.sota2.com/research/sota/video-restoration-on-udm10.md?metric=VQ-Insight%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | LPIPS | Result date | Report paper |
| --- | --- | --- | --- |
| SeedVR2 | 0.265 | Not recorded | 69c36c430636e5bf4afc66c2 |
| DOVE | 0.265 | Not recorded | 69c36c430636e5bf4afc66c2 |
| DOVER-guide | 0.276 | Not recorded | 69c36c430636e5bf4afc66c2 |
| FlashVSR | 0.288 | Not recorded | 69c36c430636e5bf4afc66c2 |
| VQ-Jarvis | 0.288 | Not recorded | 69c36c430636e5bf4afc66c2 |
| MGLD-VSR | 0.31 | Not recorded | 69c36c430636e5bf4afc66c2 |

### SeedVR2

LPIPS: 0.265. Result ID: 69c3eae047f34ce924ad9c97.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c430636e5bf4afc66c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79381e209131cea7caf5.md>)

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

### DOVE

LPIPS: 0.265. Result ID: 69c3eae047f34ce924ad9c99.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c430636e5bf4afc66c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78be1e209131cea7a325.md>)

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

### DOVER-guide

LPIPS: 0.276. Result ID: 69c3eae047f34ce924ad9c9b.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c430636e5bf4afc66c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb71e209131cea41b5c.md>)

[Implementation](<https://github.com/VQAssessment/DOVER>)

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

### FlashVSR

LPIPS: 0.288. Result ID: 69c3eae047f34ce924ad9c9a.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c430636e5bf4afc66c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1c1e209131cea8c683.md>)

[Implementation](<https://github.com/OpenImagingLab/FlashVSR>)

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

### VQ-Jarvis

LPIPS: 0.288. Result ID: 69c3eae047f34ce924ad9c9c.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c430636e5bf4afc66c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36c430636e5bf4afc66c2.md>)

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

### MGLD-VSR

LPIPS: 0.31. Result ID: 69c3eae047f34ce924ad9c98.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c430636e5bf4afc66c2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f720.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.
