# Image Manipulation Detection on StarGAN CelebA-HQ (test)

Source: [Image Manipulation Detection on StarGAN CelebA-HQ (test)](<https://www.sota2.com/research/sota/image-manipulation-detection-on-stargan-celeba-hq-test>)

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

Data updated: 2026-02-26T05:22:42.619Z

Evaluation of models designed to detect manipulated images using the test split of the StarGAN CelebA-HQ dataset, typically measured using True Detection Rate (TDR) at a specific False Alarm Rate (FAR).

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 | Image Manipulation Detection |
| Dataset (including any stated split) | StarGAN CelebA-HQ (test) |
| Metric | AP |
| Unit | PERCENT |
| Direction | Higher 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

- [AP](<https://www.sota2.com/research/sota/image-manipulation-detection-on-stargan-celeba-hq-test.md?metric=AP>): Higher is better; unit: PERCENT.

- [TDR @ 0.5% FAR](<https://www.sota2.com/research/sota/image-manipulation-detection-on-stargan-celeba-hq-test.md?metric=TDR%20%40%200.5%25%20FAR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP | Result date | Report paper |
| --- | --- | --- | --- |
| \[59\] | 100 | Not recorded | 68fa6c2f1e209131cea32a34 |
| \[48\] | 100 | Not recorded | 68fa6c2f1e209131cea32a34 |
| proactive IMD | 100 | Not recorded | 68fa6c2f1e209131cea32a34 |
| proactive IMD | 100 | Not recorded | 68fa6c2f1e209131cea32a34 |
| proactive IMD | 100 | Not recorded | 68fa6c2f1e209131cea32a34 |
| proactive IMD | 97.89 | Not recorded | 68fa6c2f1e209131cea32a34 |
| proactive IMD | 97.18 | Not recorded | 68fa6c2f1e209131cea32a34 |
| \[10\] | 91.7 | Not recorded | 68fa6c2f1e209131cea32a34 |
| \[28\] | 88.2 | Not recorded | 68fa6c2f1e209131cea32a34 |

### \[59\]

AP: 100. Result ID: 698eec653a6c314849f37970.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa667d1e209131cea108e7.md>)

Recorded evaluation configuration: {"Train GM":"AutoGAN"}. This record does not assert independent reproduction.

### \[48\]

AP: 100. Result ID: 698eec653a6c314849f37971.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67581e209131cea15b26.md>)

Recorded evaluation configuration: {"Train GM":"ProGAN"}. This record does not assert independent reproduction.

### proactive IMD

AP: 100. Result ID: 698eec653a6c314849f37972.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Implementation](<https://www.github.com/vishal3477/proactive_IMD>)

Recorded evaluation configuration: {"Train GM":"STGAN","Set size":"3"}. This record does not assert independent reproduction.

### proactive IMD

AP: 100. Result ID: 698eec653a6c314849f37973.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Implementation](<https://www.github.com/vishal3477/proactive_IMD>)

Recorded evaluation configuration: {"Train GM":"STGAN","Set size":"20"}. This record does not assert independent reproduction.

### proactive IMD

AP: 100. Result ID: 698eec653a6c314849f37976.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Implementation](<https://www.github.com/vishal3477/proactive_IMD>)

Recorded evaluation configuration: {"Train GM":"STGAN + AutoGAN","Set size":"3"}. This record does not assert independent reproduction.

### proactive IMD

AP: 97.89. Result ID: 698eec653a6c314849f37974.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Implementation](<https://www.github.com/vishal3477/proactive_IMD>)

Recorded evaluation configuration: {"Train GM":"AutoGAN","Set size":"3"}. This record does not assert independent reproduction.

### proactive IMD

AP: 97.18. Result ID: 698eec653a6c314849f37975.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Implementation](<https://www.github.com/vishal3477/proactive_IMD>)

Recorded evaluation configuration: {"Train GM":"AutoGAN","Set size":"20"}. This record does not assert independent reproduction.

### \[10\]

AP: 91.7. Result ID: 698eec653a6c314849f3796f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66b81e209131cea12171.md>)

Recorded evaluation configuration: {"Train GM":"ProGAN"}. This record does not assert independent reproduction.

### \[28\]

AP: 88.2. Result ID: 698eec653a6c314849f3796e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a34.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66751e209131cea10378.md>)

Recorded evaluation configuration: {"Train GM":"CycleGAN"}. This record does not assert independent reproduction.

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