# Face Verification on MLFW

Source: [Face Verification on MLFW](<https://www.sota2.com/research/sota/face-verification-on-mlfw>)

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

Data updated: 2026-02-26T05:28:13.646Z

Evaluation of face verification performance on the MLFW dataset, typically reported using 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 | Face Verification |
| Dataset (including any stated split) | MLFW |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/face-verification-on-mlfw.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DeepFaceEMD | 91.17 | Not recorded | 68fa6c1a1e209131cea32323 |
| Wang et al. | 90.6 | Not recorded | 68fa6c1a1e209131cea32323 |
| DeepFaceEMD | 87.17 | Not recorded | 68fa6c1a1e209131cea32323 |
| Wang et al. | 82.87 | Not recorded | 68fa6c1a1e209131cea32323 |
| DeepFaceEMD | 76.5 | Not recorded | 68fa6c1a1e209131cea32323 |
| Wang et al. | 74.85 | Not recorded | 68fa6c1a1e209131cea32323 |

### DeepFaceEMD

Accuracy: 91.17. Result ID: 698ed409ffc3901de0a1ce4c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

[Implementation](<https://github.com/anguyen8/deepface-emd>)

Recorded evaluation configuration: {"Training Dataset":"MS1MV2","Backbone":"R100","Loss Function":"Curricularface","Similarity Function":"DeepFaceEMD"}. This record does not assert independent reproduction.

### Wang et al.

Accuracy: 90.6. Result ID: 698ed409ffc3901de0a1ce4b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

Recorded evaluation configuration: {"Training Dataset":"MS1MV2","Backbone":"R100","Loss Function":"Curricularface","Similarity Function":"Baseline"}. This record does not assert independent reproduction.

### DeepFaceEMD

Accuracy: 87.17. Result ID: 698ed409ffc3901de0a1ce4a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

[Implementation](<https://github.com/anguyen8/deepface-emd>)

Recorded evaluation configuration: {"Training Dataset":"CASIA","Backbone":"R50","Loss Function":"CosFace","Similarity Function":"DeepFaceEMD"}. This record does not assert independent reproduction.

### Wang et al.

Accuracy: 82.87. Result ID: 698ed409ffc3901de0a1ce49.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

Recorded evaluation configuration: {"Training Dataset":"CASIA","Backbone":"R50","Loss Function":"CosFace","Similarity Function":"Baseline"}. This record does not assert independent reproduction.

### DeepFaceEMD

Accuracy: 76.5. Result ID: 698ed409ffc3901de0a1ce48.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

[Implementation](<https://github.com/anguyen8/deepface-emd>)

Recorded evaluation configuration: {"Training Dataset":"Private-Asia","Backbone":"R50","Loss Function":"ArcFace","Similarity Function":"DeepFaceEMD"}. This record does not assert independent reproduction.

### Wang et al.

Accuracy: 74.85. Result ID: 698ed409ffc3901de0a1ce47.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c1a1e209131cea32323.md>)

Recorded evaluation configuration: {"Training Dataset":"Private-Asia","Backbone":"R50","Loss Function":"ArcFace","Similarity Function":"Baseline"}. This record does not assert independent reproduction.

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