# Random Inpainting on ImageNet

Source: [Random Inpainting on ImageNet](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet>)

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

Data updated: 2026-05-08T07:53:12.922Z

Evaluation of image inpainting performance on the ImageNet dataset using random masking strategies, measured by PSNR, SSIM, LPIPS, FID, MMD, CRPS metrics, and inference time.

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 | Random Inpainting |
| Dataset (including any stated split) | ImageNet |
| Metric | LPIPS |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 11 of 11 |
| 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/random-inpainting-on-imagenet.md?metric=LPIPS>): Lower is better; unit: SCALAR.

- [PSNR](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet.md?metric=PSNR>): Higher is better; unit: SCALAR.

- [SSIM](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet.md?metric=SSIM>): Higher is better; unit: PERCENT.

- [FID](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet.md?metric=FID>): Lower is better; unit: SCALAR.

- [MMD](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet.md?metric=MMD>): Lower is better; unit: SCALAR.

- [CRPS (DINO)](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet.md?metric=CRPS%20(DINO)>): Lower is better; unit: SCALAR.

- [CRPS (Inception)](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet.md?metric=CRPS%20(Inception)>): Lower is better; unit: SCALAR.

- [Inference Time (s)](<https://www.sota2.com/research/sota/random-inpainting-on-imagenet.md?metric=Inference%20Time%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | LPIPS | Result date | Report paper |
| --- | --- | --- | --- |
| LCDM-BAOAB | 0.0933 | Not recorded | 69fd6de75497b3b670c89293 |
| DDNM | 0.1182 | Not recorded | 69fd6de75497b3b670c89293 |
| VFM | 0.336 | Not recorded | 69afa5702d603e6de3e43970 |
| Latent DPS | 0.337 | Not recorded | 69afa5702d603e6de3e43970 |
| PSLD | 0.338 | Not recorded | 69afa5702d603e6de3e43970 |
| FlowChef | 0.338 | Not recorded | 69afa5702d603e6de3e43970 |
| MPGD | 0.339 | Not recorded | 69afa5702d603e6de3e43970 |
| FlowDPS | 0.344 | Not recorded | 69afa5702d603e6de3e43970 |
| VFM | 0.367 | Not recorded | 69afa5702d603e6de3e43970 |
| Latent DAPS | 0.384 | Not recorded | 69afa5702d603e6de3e43970 |
| frozen-θ | 0.53 | Not recorded | 69afa5702d603e6de3e43970 |

### LCDM-BAOAB

LPIPS: 0.0933. Result ID: 69fd966447f34ce924e05e0a.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de75497b3b670c89293.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6de75497b3b670c89293.md>)

Recorded evaluation configuration: {"Mask size":"100x100","Resolution":"256x256","Sample size":"1000"}. This record does not assert independent reproduction.

### DDNM

LPIPS: 0.1182. Result ID: 69fd966447f34ce924e05e09.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de75497b3b670c89293.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e201e209131cea3dfb9.md>)

[Implementation](<https://github.com/wyhuai/DDNM>)

Recorded evaluation configuration: {"Mask size":"100x100","Resolution":"256x256","Sample size":"1000"}. This record does not assert independent reproduction.

### VFM

LPIPS: 0.336. Result ID: 69afdeaa6dbb7075b9586803.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Implementation](<https://github.com/abbasmammadov/VFM>)

Recorded evaluation configuration: {"NFE":"10"}. This record does not assert independent reproduction.

### Latent DPS

LPIPS: 0.337. Result ID: 69afdeaa6dbb7075b95867fb.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bdd1e209131cea8adbc.md>)

Recorded evaluation configuration: {"NFE":"250×2"}. This record does not assert independent reproduction.

### PSLD

LPIPS: 0.338. Result ID: 69afdeaa6dbb7075b95867fd.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fe71e209131cea48057.md>)

[Implementation](<https://github.com/LituRout/PSLD>)

Recorded evaluation configuration: {"NFE":"250×2"}. This record does not assert independent reproduction.

### FlowChef

LPIPS: 0.338. Result ID: 69afdeaa6dbb7075b95867ff.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

Recorded evaluation configuration: {"NFE":"250×2"}. This record does not assert independent reproduction.

### MPGD

LPIPS: 0.339. Result ID: 69afdeaa6dbb7075b95867fe.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71461e209131cea504ca.md>)

Recorded evaluation configuration: {"NFE":"250×2"}. This record does not assert independent reproduction.

### FlowDPS

LPIPS: 0.344. Result ID: 69afdeaa6dbb7075b9586800.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77641e209131cea72f7d.md>)

Recorded evaluation configuration: {"NFE":"250×2"}. This record does not assert independent reproduction.

### VFM

LPIPS: 0.367. Result ID: 69afdeaa6dbb7075b9586802.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Implementation](<https://github.com/abbasmammadov/VFM>)

Recorded evaluation configuration: {"NFE":"1"}. This record does not assert independent reproduction.

### Latent DAPS

LPIPS: 0.384. Result ID: 69afdeaa6dbb7075b95867fc.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7aa31e209131cea84481.md>)

Recorded evaluation configuration: {"NFE":"250×2"}. This record does not assert independent reproduction.

### frozen-θ

LPIPS: 0.53. Result ID: 69afdeaa6dbb7075b9586801.

[Reporting paper](<https://www.sota2.com/research/paper/69afa5702d603e6de3e43970.md>)

Recorded evaluation configuration: {"NFE":"1"}. This record does not assert independent reproduction.

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