Facial Landmark Detection on MAFL (test)
2.42Normalised MSE (%)Deep Latent Particles
Evaluation Results
| Method | Links | |
|---|---|---|
| Deep Latent ParticlesK=50, features=mean + log-variance2022.05 | 2.42 | |
| Deep Latent ParticlesK=50, features=mean2022.05 | 2.43 | |
| Deep Latent ParticlesK=30, features=mean + log-variance2022.05 | 2.49 | |
| Deep Latent ParticlesK=25, features=mean + log-variance2022.05 | 2.52 | |
| Proposed MethodTraining Set=CelebA, Loss Network=self-supervised, K=502018.06 | 2.54 | |
| KeyNetK=502022.05 | 2.54 | |
| Deep Latent ParticlesK=30, features=mean2022.05 | 2.56 | |
| Proposed MethodTraining Set=CelebA, Loss Network=self-supervised, K=302018.06 | 2.58 | |
| KeyNetK=302022.05 | 2.58 | |
| Proposed MethodTraining Set=CelebA, Loss Network=supervised, K=502018.06 | 2.59 | |
| Proposed MethodTraining Set=CelebA, Loss Network=supervised, K=302018.06 | 2.63 | |
| Deep Latent ParticlesK=25, features=mean2022.05 | 2.87 | |
| ZhangSupervision=Unsupervised, K=30, Equivariance=true2018.06 | 3.16 | |
| ZhangK=302022.05 | 3.16 | |
| Proposed MethodTraining Set=CelebA, Loss Network=self-supervised, K=102018.06 | 3.19 | |
| Proposed MethodTraining Set=CelebA, Loss Network=supervised, K=102018.06 | 3.32 | |
| WilesSupervision=Unsupervised, Dataset=VoxCeleb2, Regressor Bias=true2018.06 | 3.44 | |
| ZhangSupervision=Unsupervised, K=102018.06 | 3.46 | |
| Proposed MethodTraining Set=VoxCeleb, Loss Network=self-supervised, K=30, Regressor Bias=true2018.06 | 3.63 | |
| Proposed MethodTraining Set=VoxCeleb, Loss Network=self-supervised, K=302018.06 | 3.94 | |
| Proposed MethodTraining Set=VoxCeleb, Loss Network=supervised, K=302018.06 | 4.01 | |
| MTCNNSupervision=Supervised2018.06 | 5.39 | |
| Shu +Supervision=Unsupervised2018.06 | 5.45 | |
| Thewlis (frames)Supervision=Unsupervised2018.06 | 5.83 | |
| ThewlisSupervision=Unsupervised, K=502018.06 | 6.67 | |
| ThewlisSupervision=Unsupervised, K=302018.06 | 7.15 | |
| TCDCNSupervision=Supervised2018.06 | 7.95 | |
| ZhangSupervision=Unsupervised, K=30, Equivariance=false2018.06 | 8.42 | |
| Cascaded CNNSupervision=Supervised2018.06 | 9.73 | |
| CFANSupervision=Supervised2018.06 | 15.84 |