# Full-image colorization on ImageNet (ctest10k)

Source: [Full-image colorization on ImageNet (ctest10k)](<https://www.sota2.com/research/sota/full-image-colorization-on-imagenet-ctest10k>)

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

Data updated: 2026-02-26T05:48:11.520Z

Evaluation of full-image colorization performance on the ctest10k subset of the ImageNet dataset, measured using LPIPS, 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 | Full-image colorization |
| Dataset (including any stated split) | ImageNet (ctest10k) |
| Metric | LPIPS |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 9 of 9 |
| 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/full-image-colorization-on-imagenet-ctest10k.md?metric=LPIPS>): Lower is better; unit: SCALAR.

- [PSNR](<https://www.sota2.com/research/sota/full-image-colorization-on-imagenet-ctest10k.md?metric=PSNR>): Higher is better; unit: SCALAR.

- [SSIM](<https://www.sota2.com/research/sota/full-image-colorization-on-imagenet-ctest10k.md?metric=SSIM>): Higher is better; unit: PERCENT.

## Recorded results

| Method | LPIPS | Result date | Report paper |
| --- | --- | --- | --- |
| Instance-aware Image Colorization | 0.125 | Not recorded | 68fa67971e209131cea17460 |
| Instance-aware Image Colorization | 0.134 | Not recorded | 68fa67971e209131cea17460 |
| Zhang et al. \[41\] | 0.14 | Not recorded | 68fa67971e209131cea17460 |
| Zhang et al. \[41\] | 0.145 | Not recorded | 68fa67971e209131cea17460 |
| DeOldify et al. | 0.187 | Not recorded | 68fa67971e209131cea17460 |
| Larsson et al. | 0.188 | Not recorded | 68fa67971e209131cea17460 |
| Iizuka et al. | 0.2 | Not recorded | 68fa67971e209131cea17460 |
| Lei et al. | 0.202 | Not recorded | 68fa67971e209131cea17460 |
| Zhang et al. \[38\] | 0.238 | Not recorded | 68fa67971e209131cea17460 |

### Instance-aware Image Colorization

LPIPS: 0.125. Result ID: 698d3cae8b517eb4331ffe40.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Implementation](<https://ericsujw.github.io/InstColorization>)

Recorded evaluation configuration: {"Finetuned on COCO-Stuff":true}. This record does not assert independent reproduction.

### Instance-aware Image Colorization

LPIPS: 0.134. Result ID: 698d3cae8b517eb4331ffe3e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Implementation](<https://ericsujw.github.io/InstColorization>)

Recorded evaluation configuration: {"Training Dataset":"ImageNet"}. This record does not assert independent reproduction.

### Zhang et al. \[41\]

LPIPS: 0.14. Result ID: 698d3cae8b517eb4331ffe3f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63531e209131ce9fe351.md>)

[Implementation](<https://richzhang.github.io/ideepcolor>)

Recorded evaluation configuration: {"Finetuned on COCO-Stuff":true}. This record does not assert independent reproduction.

### Zhang et al. \[41\]

LPIPS: 0.145. Result ID: 698d3cae8b517eb4331ffe3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63531e209131ce9fe351.md>)

[Implementation](<https://richzhang.github.io/ideepcolor>)

Recorded evaluation configuration: {"Training Dataset":"ImageNet"}. This record does not assert independent reproduction.

### DeOldify et al.

LPIPS: 0.187. Result ID: 698d3cae8b517eb4331ffe3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

Recorded evaluation configuration: {"Training Dataset":"ImageNet"}. This record does not assert independent reproduction.

### Larsson et al.

LPIPS: 0.188. Result ID: 698d3cae8b517eb4331ffe39.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63881e209131ce9ff4ff.md>)

Recorded evaluation configuration: {"Training Dataset":"ImageNet"}. This record does not assert independent reproduction.

### Iizuka et al.

LPIPS: 0.2. Result ID: 698d3cae8b517eb4331ffe38.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

Recorded evaluation configuration: {"Training Dataset":"ImageNet"}. This record does not assert independent reproduction.

### Lei et al.

LPIPS: 0.202. Result ID: 698d3cae8b517eb4331ffe3d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa661a1e209131cea0e4de.md>)

Recorded evaluation configuration: {"Training Dataset":"ImageNet"}. This record does not assert independent reproduction.

### Zhang et al. \[38\]

LPIPS: 0.238. Result ID: 698d3cae8b517eb4331ffe3a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67971e209131cea17460.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62d41e209131ce9fb63e.md>)

[Implementation](<http://richzhang.github.io/colorization/>)

Recorded evaluation configuration: {"Training Dataset":"ImageNet"}. This record does not assert independent reproduction.

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