# Face Recognition on LFW, CA-LFW, CP-LFW, CFP-FP, and AgeDB-30 (test) Source: [Face Recognition on LFW, CA-LFW, CP-LFW, CFP-FP, and AgeDB-30 (test)]() Collection index: [llms.txt]() Data updated: 2026-04-13T09:05:34.355Z Aggregate evaluation of face recognition performance across the test splits of five standard datasets: LFW, CA-LFW, CP-LFW, CFP-FP, and AgeDB-30, including accuracy and system latency metrics. 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 Recognition | | Dataset (including any stated split) | LFW, CA-LFW, CP-LFW, CFP-FP, and AgeDB-30 (test) | | Metric | Mean Accuracy (%) | | Unit | PERCENT | | 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 - [Mean Accuracy (%)](): Higher is better; unit: PERCENT. - [ONNX GPU Latency (ms)](): Lower is better; unit: TIME. - [ONNX CPU Latency (ms)](): Lower is better; unit: TIME. - [ONNX Jetson GPU Latency (ms)](): Lower is better; unit: TIME. - [ONNX Jetson CPU Latency (ms)](): Lower is better; unit: TIME. ## Recorded results | Method | Mean Accuracy (%) | Result date | Report paper | | --- | --- | --- | --- | | FaceLiVTv2-L | 97.03 | Not recorded | 69dc78697d0172140ec78986 | | FaceLiVTv2-M | 96.97 | Not recorded | 69dc78697d0172140ec78986 | | FaceLiVTv2-S | 96.69 | Not recorded | 69dc78697d0172140ec78986 | | KANFace-0.5 | 96.63 | Not recorded | 69dc78697d0172140ec78986 | | GhostFaceNetV1-1 | 96.48 | Not recorded | 69dc78697d0172140ec78986 | | FaceLiVTv1-M | 96.25 | Not recorded | 69dc78697d0172140ec78986 | | EdgeFace-S | 96.16 | Not recorded | 69dc78697d0172140ec78986 | | SwiftFaceFormer-L1 | 95.83 | Not recorded | 69dc78697d0172140ec78986 | | KANFace-0.6 | 95.83 | Not recorded | 69dc78697d0172140ec78986 | | SwiftFaceFormer-S | 95.74 | Not recorded | 69dc78697d0172140ec78986 | | FaceLiVTv1-S | 95.55 | Not recorded | 69dc78697d0172140ec78986 | | FaceLiVTv2-XS | 95.5 | Not recorded | 69dc78697d0172140ec78986 | | EdgeFace-XS | 95.44 | Not recorded | 69dc78697d0172140ec78986 | | GhostFaceNetV2-2 | 95.34 | Not recorded | 69dc78697d0172140ec78986 | | GhostFaceNetV1-2 | 95.12 | Not recorded | 69dc78697d0172140ec78986 | | SwiftFaceFormer-XS | 95.08 | Not recorded | 69dc78697d0172140ec78986 | ### FaceLiVTv2-L Mean Accuracy (%): 97.03. Result ID: 69dcb1c947f34ce924c3d35e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### FaceLiVTv2-M Mean Accuracy (%): 96.97. Result ID: 69dcb1c947f34ce924c3d35f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### FaceLiVTv2-S Mean Accuracy (%): 96.69. Result ID: 69dcb1c947f34ce924c3d360. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### KANFace-0.5 Mean Accuracy (%): 96.63. Result ID: 69dcb1c947f34ce924c3d352. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### GhostFaceNetV1-1 Mean Accuracy (%): 96.48. Result ID: 69dcb1c947f34ce924c3d357. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### FaceLiVTv1-M Mean Accuracy (%): 96.25. Result ID: 69dcb1c947f34ce924c3d353. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### EdgeFace-S Mean Accuracy (%): 96.16. Result ID: 69dcb1c947f34ce924c3d354. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### SwiftFaceFormer-L1 Mean Accuracy (%): 95.83. Result ID: 69dcb1c947f34ce924c3d355. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### KANFace-0.6 Mean Accuracy (%): 95.83. Result ID: 69dcb1c947f34ce924c3d358. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### SwiftFaceFormer-S Mean Accuracy (%): 95.74. Result ID: 69dcb1c947f34ce924c3d356. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### FaceLiVTv1-S Mean Accuracy (%): 95.55. Result ID: 69dcb1c947f34ce924c3d359. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### FaceLiVTv2-XS Mean Accuracy (%): 95.5. Result ID: 69dcb1c947f34ce924c3d361. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### EdgeFace-XS Mean Accuracy (%): 95.44. Result ID: 69dcb1c947f34ce924c3d35a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### GhostFaceNetV2-2 Mean Accuracy (%): 95.34. Result ID: 69dcb1c947f34ce924c3d35c. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### GhostFaceNetV1-2 Mean Accuracy (%): 95.12. Result ID: 69dcb1c947f34ce924c3d35d. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. ### SwiftFaceFormer-XS Mean Accuracy (%): 95.08. Result ID: 69dcb1c947f34ce924c3d35b. [Reporting paper]() Recorded evaluation configuration: {"Runtime":"ONNX"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.