Face Verification On Megaface
评估指标
Accuracy
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | ||
|---|---|---|---|
| Prodpoly | 98.95% | Deep Polynomial Neural Networks | |
| ElasticFace-Arc | 98.81% | ElasticFace: Elastic Margin Loss for Deep Face Recognition | |
| GhostFaceNetV2-1 | 98.72% | GhostFaceNets: Lightweight Face Recognition Model From Cheap Operations | - |
| ArcFace + MS1MV2 + R100 + R | 98.48% | ArcFace: Additive Angular Margin Loss for Deep Face Recognition | |
| DiscFace | 97.44% | DiscFace: Minimum Discrepancy Learning for Deep Face Recognition | - |
| Dynamic AdaCos | 97.41% | AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations | |
| SV-AM-Softmax | 97.38% | Support Vector Guided Softmax Loss for Face Recognition | |
| CosFace | 96.65% | CosFace: Large Margin Cosine Loss for Deep Face Recognition | |
| PFEfuse + match | 92.51% | Probabilistic Face Embeddings | |
| SphereFace (3-patch ensemble) | 89.142% | SphereFace: Deep Hypersphere Embedding for Face Recognition | |
| SphereFace (single model) | 85.561% | SphereFace: Deep Hypersphere Embedding for Face Recognition | |
| Light CNN-29 | 85.133% | A Light CNN for Deep Face Representation with Noisy Labels |
0 of 12 row(s) selected.