# Image Watermarking Evaluation on MS-COCO 2014 (val)

Source: [Image Watermarking Evaluation on MS-COCO 2014 (val)](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val>)

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

Data updated: 2026-02-26T05:41:17.692Z

Evaluation of various Image Watermarking algorithms (e.g., DwtDct, RivaGAN, HiDDeN) on the validation split of the MS-COCO 2014 dataset, focusing on robustness (PLCC/SRCC) and security accuracy.

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 Watermarking Evaluation |
| Dataset (including any stated split) | MS-COCO 2014 (val) |
| Metric | DwtDct PLCC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [DwtDct PLCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=DwtDct%20PLCC>): Higher is better; unit: SCALAR.

- [DwtDct SRCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=DwtDct%20SRCC>): Higher is better; unit: SCALAR.

- [DwtDct Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=DwtDct%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [RivaGAN PLCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RivaGAN%20PLCC>): Higher is better; unit: SCALAR.

- [RivaGAN SRCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RivaGAN%20SRCC>): Higher is better; unit: SCALAR.

- [RivaGAN Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RivaGAN%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [HiDDeN PLCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=HiDDeN%20PLCC>): Higher is better; unit: SCALAR.

- [HiDDeN SRCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=HiDDeN%20SRCC>): Higher is better; unit: SCALAR.

- [HiDDeN Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=HiDDeN%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [RW PLCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RW%20PLCC>): Higher is better; unit: SCALAR.

- [RW SRCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RW%20SRCC>): Higher is better; unit: SCALAR.

- [RW Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RW%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [VINE PLCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=VINE%20PLCC>): Higher is better; unit: SCALAR.

- [VINE SRCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=VINE%20SRCC>): Higher is better; unit: SCALAR.

- [VINE Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=VINE%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [SS PLCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=SS%20PLCC>): Higher is better; unit: SCALAR.

- [SS SRCC](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=SS%20SRCC>): Higher is better; unit: SCALAR.

- [SS Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=SS%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [RingID Quality Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RingID%20Quality%20Accuracy>): Higher is better; unit: SCALAR.

- [RingID Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=RingID%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [Tree-Ring Quality Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=Tree-Ring%20Quality%20Accuracy>): Higher is better; unit: SCALAR.

- [Tree-Ring Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=Tree-Ring%20Security%20Accuracy>): Higher is better; unit: SCALAR.

- [Lossless Quality Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=Lossless%20Quality%20Accuracy>): Higher is better; unit: SCALAR.

- [Lossless Security Accuracy](<https://www.sota2.com/research/sota/image-watermarking-evaluation-on-ms-coco-2014-val.md?metric=Lossless%20Security%20Accuracy>): Higher is better; unit: SCALAR.

## Recorded results

| Method | DwtDct PLCC | Result date | Report paper |
| --- | --- | --- | --- |
| WMVLM | 0.933 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Qwen3-VL-8B | 0.198 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Qwen3-VL-32B | 0.114 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Gemma-3-12B | 0.111 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Gemma-3-27B | 0.083 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Gemma-3-4B | 0.043 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Gemini-3-Pro | 0.007 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Claude-Opus-4.5 | -0.06 | Not recorded | 697c3b54a8b01e76cc8da60e |
| GPT-5 | -0.089 | Not recorded | 697c3b54a8b01e76cc8da60e |
| LLaVA-v1.6-7B | -0.255 | Not recorded | 697c3b54a8b01e76cc8da60e |
| Qwen3-VL-4B | -0.305 | Not recorded | 697c3b54a8b01e76cc8da60e |
| LLaVA-v1.6-13B | -0.316 | Not recorded | 697c3b54a8b01e76cc8da60e |

### WMVLM

DwtDct PLCC: 0.933. Result ID: 69977aca5698665ec4cf27e8.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

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

### Qwen3-VL-8B

DwtDct PLCC: 0.198. Result ID: 69977aca5698665ec4cf27e3.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### Qwen3-VL-32B

DwtDct PLCC: 0.114. Result ID: 69977aca5698665ec4cf27e4.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Parameters":"32B"}. This record does not assert independent reproduction.

### Gemma-3-12B

DwtDct PLCC: 0.111. Result ID: 69977aca5698665ec4cf27e6.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

Recorded evaluation configuration: {"Parameters":"12B"}. This record does not assert independent reproduction.

### Gemma-3-27B

DwtDct PLCC: 0.083. Result ID: 69977aca5698665ec4cf27e7.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

Recorded evaluation configuration: {"Parameters":"27B"}. This record does not assert independent reproduction.

### Gemma-3-4B

DwtDct PLCC: 0.043. Result ID: 69977aca5698665ec4cf27e5.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction.

### Gemini-3-Pro

DwtDct PLCC: 0.007. Result ID: 69977aca5698665ec4cf27df.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

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

### Claude-Opus-4.5

DwtDct PLCC: -0.06. Result ID: 69977aca5698665ec4cf27de.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

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

### GPT-5

DwtDct PLCC: -0.089. Result ID: 69977aca5698665ec4cf27dd.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

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

### LLaVA-v1.6-7B

DwtDct PLCC: -0.255. Result ID: 69977aca5698665ec4cf27e0.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction.

### Qwen3-VL-4B

DwtDct PLCC: -0.305. Result ID: 69977aca5698665ec4cf27e2.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction.

### LLaVA-v1.6-13B

DwtDct PLCC: -0.316. Result ID: 69977aca5698665ec4cf27e1.

[Reporting paper](<https://www.sota2.com/research/paper/697c3b54a8b01e76cc8da60e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Parameters":"13B"}. This record does not assert independent reproduction.

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