# Motion blur removal on Non-uniform motion blur dataset Large Blur

Source: [Motion blur removal on Non-uniform motion blur dataset Large Blur](<https://www.sota2.com/research/sota/motion-blur-removal-on-non-uniform-motion-blur-dataset-large-blur>)

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

Data updated: 2026-02-26T05:55:22.617Z

Evaluation of motion blur removal performance on the Large Blur subset of the Non-uniform motion blur dataset, typically measured using PSNR and SSIM.

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 | Motion blur removal |
| Dataset (including any stated split) | Non-uniform motion blur dataset Large Blur |
| Metric | PSNR |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [PSNR](<https://www.sota2.com/research/sota/motion-blur-removal-on-non-uniform-motion-blur-dataset-large-blur.md?metric=PSNR>): Higher is better; unit: SCALAR.

- [SSIM](<https://www.sota2.com/research/sota/motion-blur-removal-on-non-uniform-motion-blur-dataset-large-blur.md?metric=SSIM>): Higher is better; unit: SCALAR.

## Recorded results

| Method | PSNR | Result date | Report paper |
| --- | --- | --- | --- |
| PFF | 22.3 | Not recorded | 68fa651e1e209131cea087fd |
| Bahat et al. | 22.01 | Not recorded | 68fa651e1e209131cea087fd |
| Our CNN | 21.06 | Not recorded | 68fa651e1e209131cea087fd |
| Sun et al. | 20.84 | Not recorded | 68fa651e1e209131cea087fd |
| \[55\] | 20.47 | Not recorded | 68fa651e1e209131cea087fd |

### PFF

PSNR: 22.3. Result ID: 698c9e960b614d12cff2017c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa651e1e209131cea087fd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa651e1e209131cea087fd.md>)

[Implementation](<https://github.com/aimerykong/predictive-filter-flow>)

Recorded evaluation configuration: {"predicted filter size":"17x17"}. This record does not assert independent reproduction.

### Bahat et al.

PSNR: 22.01. Result ID: 698c9e960b614d12cff2017a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa651e1e209131cea087fd.md>)

Recorded evaluation configuration: {"mode":"patch-optim"}. This record does not assert independent reproduction.

### Our CNN

PSNR: 21.06. Result ID: 698c9e960b614d12cff2017b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa651e1e209131cea087fd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa651e1e209131cea087fd.md>)

[Implementation](<https://github.com/aimerykong/predictive-filter-flow>)

Recorded evaluation configuration: {"architecture":"two-stream"}. This record does not assert independent reproduction.

### Sun et al.

PSNR: 20.84. Result ID: 698c9e960b614d12cff20179.

[Reporting paper](<https://www.sota2.com/research/paper/68fa651e1e209131cea087fd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61d01e209131ce9f6347.md>)

Recorded evaluation configuration: {"backbone":"CNN"}. This record does not assert independent reproduction.

### \[55\]

PSNR: 20.47. Result ID: 698c9e960b614d12cff20178.

[Reporting paper](<https://www.sota2.com/research/paper/68fa651e1e209131cea087fd.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.