# Image-level manipulation detection on CASIA v1

Source: [Image-level manipulation detection on CASIA v1](<https://www.sota2.com/research/sota/image-level-manipulation-detection-on-casia-v1>)

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

Data updated: 2026-05-14T07:16:43.679Z

Evaluation of Image-level manipulation detection performance on the CASIA v1 dataset.

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-level manipulation detection |
| Dataset (including any stated split) | CASIA v1 |
| Metric | AUC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [AUC](<https://www.sota2.com/research/sota/image-level-manipulation-detection-on-casia-v1.md?metric=AUC>): Higher is better; unit: SCALAR.

- [Sensitivity](<https://www.sota2.com/research/sota/image-level-manipulation-detection-on-casia-v1.md?metric=Sensitivity>): Higher is better; unit: SCALAR.

- [Specificity](<https://www.sota2.com/research/sota/image-level-manipulation-detection-on-casia-v1.md?metric=Specificity>): Higher is better; unit: SCALAR.

- [F1 Score](<https://www.sota2.com/research/sota/image-level-manipulation-detection-on-casia-v1.md?metric=F1%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | AUC | Result date | Report paper |
| --- | --- | --- | --- |
| MVSS-Net | 0.839 | Not recorded | 68fa6a751e209131cea28443 |
| FCN | 0.796 | Not recorded | 68fa6a751e209131cea28443 |
| CR-CNN | 0.766 | Not recorded | 68fa6a751e209131cea28443 |
| FRAME | 0.741 | Not recorded | 6a05571edbc836b87d828d3c |
| TruFor | 0.724 | Not recorded | 6a05571edbc836b87d828d3c |
| CAT-Net | 0.678 | Not recorded | 6a05571edbc836b87d828d3c |
| XGB-Ensemble (pyIFD) | 0.674 | Not recorded | 6a05571edbc836b87d828d3c |
| RF-Ensemble (pyIFD) | 0.658 | Not recorded | 6a05571edbc836b87d828d3c |
| ManTraNet | 0.651 | Not recorded | 6a05571edbc836b87d828d3c |
| Heuristic-K + uniform | 0.628 | Not recorded | 6a05571edbc836b87d828d3c |
| Best single pyIFD | 0.612 | Not recorded | 6a05571edbc836b87d828d3c |
| Random-K + uniform | 0.589 | Not recorded | 6a05571edbc836b87d828d3c |
| Uniform-all pyIFD | 0.574 | Not recorded | 6a05571edbc836b87d828d3c |
| MMFusion | 0.51 | Not recorded | 6a05571edbc836b87d828d3c |
| GSR-Net | 0.502 | Not recorded | 68fa6a751e209131cea28443 |
| ManrTra-Net | 0.141 | Not recorded | 68fa6a751e209131cea28443 |

### MVSS-Net

AUC: 0.839. Result ID: 698e377fffc3901de09bbca1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a751e209131cea28443.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a751e209131cea28443.md>)

[Implementation](<https://github.com/dong03/MVSS-Net>)

Recorded evaluation configuration: {"Decision threshold":"0.5"}. This record does not assert independent reproduction.

### FCN

AUC: 0.796. Result ID: 698e377fffc3901de09bbca0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a751e209131cea28443.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa627b1e209131ce9f9766.md>)

[Implementation](<http://fcn.berkeleyvision.org/>)

Recorded evaluation configuration: {"Decision threshold":"0.5"}. This record does not assert independent reproduction.

### CR-CNN

AUC: 0.766. Result ID: 698e377fffc3901de09bbc9e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a751e209131cea28443.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67331e209131cea14ae7.md>)

[Implementation](<https://github.com/HuizhouLi/Constrained-R-CNN>)

Recorded evaluation configuration: {"Decision threshold":"0.5"}. This record does not assert independent reproduction.

### FRAME

AUC: 0.741. Result ID: 6a0576d347f34ce924ec062e.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

Recorded evaluation configuration: {"Selection":"Learned","Fusion":"Learned"}. This record does not assert independent reproduction.

### TruFor

AUC: 0.724. Result ID: 6a0576d347f34ce924ec062a.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f7f1e209131cea45fb0.md>)

[Implementation](<https://grip-unina.github.io/TruFor/>)

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

### CAT-Net

AUC: 0.678. Result ID: 6a0576d347f34ce924ec062d.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

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

### XGB-Ensemble (pyIFD)

AUC: 0.674. Result ID: 6a0576d347f34ce924ec0629.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f101e209131cea43a30.md>)

Recorded evaluation configuration: {"Fusion":"XGB-Ensemble"}. This record does not assert independent reproduction.

### RF-Ensemble (pyIFD)

AUC: 0.658. Result ID: 6a0576d347f34ce924ec0628.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f101e209131cea43a30.md>)

Recorded evaluation configuration: {"Fusion":"RF-Ensemble"}. This record does not assert independent reproduction.

### ManTraNet

AUC: 0.651. Result ID: 6a0576d347f34ce924ec062c.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

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

### Heuristic-K + uniform

AUC: 0.628. Result ID: 6a0576d347f34ce924ec0627.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f101e209131cea43a30.md>)

Recorded evaluation configuration: {"Selection":"Heuristic","Fusion":"Uniform"}. This record does not assert independent reproduction.

### Best single pyIFD

AUC: 0.612. Result ID: 6a0576d347f34ce924ec0624.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f101e209131cea43a30.md>)

Recorded evaluation configuration: {"Selection":"Best single"}. This record does not assert independent reproduction.

### Random-K + uniform

AUC: 0.589. Result ID: 6a0576d347f34ce924ec0626.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f101e209131cea43a30.md>)

Recorded evaluation configuration: {"Selection":"Random","Fusion":"Uniform"}. This record does not assert independent reproduction.

### Uniform-all pyIFD

AUC: 0.574. Result ID: 6a0576d347f34ce924ec0625.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f101e209131cea43a30.md>)

Recorded evaluation configuration: {"Selection":"None (all)","Fusion":"Uniform"}. This record does not assert independent reproduction.

### MMFusion

AUC: 0.51. Result ID: 6a0576d347f34ce924ec062b.

[Reporting paper](<https://www.sota2.com/research/paper/6a05571edbc836b87d828d3c.md>)

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

### GSR-Net

AUC: 0.502. Result ID: 698e377fffc3901de09bbc9f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a751e209131cea28443.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66b41e209131cea11fa8.md>)

Recorded evaluation configuration: {"Decision threshold":"0.5"}. This record does not assert independent reproduction.

### ManrTra-Net

AUC: 0.141. Result ID: 698e377fffc3901de09bbc9d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a751e209131cea28443.md>)

Recorded evaluation configuration: {"Decision threshold":"0.5"}. This record does not assert independent reproduction.

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