Face Alignment On Aflw2000 3D
Metrics
Balanced NME (2D Sparse Alignment)
Mean NME(3D Dense Alignment)
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Balanced NME (2D Sparse Alignment) | Mean NME(3D Dense Alignment) |
|---|---|---|
| face-alignment-in-full-pose-range-a-3d-total | 3.79% | 6.55% |
| adversarial-learning-semantic-volume-for-2d | 3.31% | - |
| on-the-power-of-data-augmentation-for-head | 3.55% | - |
| synergy-between-3dmm-and-3d-landmarks-for | 2.65% | - |
| joint-3d-face-reconstruction-and-dense | 3.62% | 4.40% |
| sadrnet-self-aligned-dual-face-regression | 3.46% | 4.02% |
| joint-3d-face-reconstruction-and-dense-face | 3.53% | - |
| face-alignment-across-large-poses-a-3d | 4.94% | - |
| multi-task-head-pose-estimation-in-the-wild-1 | 2.58% | - |
| dsfnet-dual-space-fusion-network-for-1 | - | 3.8% |
| faster-than-real-time-facial-alignment-a-3d | 4.49% | - |
| synergy-between-3dmm-and-3d-landmarks-for | 3.41% | 4.06% |
| dense-face-alignment | 4.50% | 6.04% |
| towards-fast-accurate-and-stable-3d-dense-1 | 3.51% | 4.18% |