3D Pose Estimation on AFLW (test)
5.324MAEMulti-Loss ResNet50
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Multi-Loss ResNet50alpha=12017.10 | 5.324 | 6.26 | 5.89 | 3.82 | — | — | — | — | — | — | |
| KEPLER2017.02 | 6.45 | 6.45 | 5.85 | 8.75 | — | — | — | — | — | — | |
| KEPLER2017.10 | 7.017 | 6.45 | 5.85 | 8.75 | — | — | — | — | — | — | |
| KEPLERsupervision=Supervised2018.08 | 7.02 | 6.45 | 5.85 | 8.75 | — | — | — | — | — | — | |
| Multi-Loss AlexNetalpha=12017.10 | 7.084 | 7.79 | 7.41 | 6.05 | — | — | — | — | — | — | |
| Patacchiola, Cangelosi2017.10 | 7.53 | 11.04 | 7.15 | 4.4 | — | — | — | — | — | — | |
| FAb-Net w/ curric., 3 source framessupervision=Self-supervised, curriculum=true, source_frames=32018.08 | 7.65 | 10.7 | 7.13 | 5.14 | — | — | — | — | — | — | |
| FAb-Net w/ curric.supervision=Self-supervised, curriculum=true2018.08 | 7.96 | 11.34 | 7.21 | 5.33 | — | — | — | — | — | — | |
| FAb-Netsupervision=Self-supervised2018.08 | 8.77 | 12.93 | 7.84 | 5.54 | — | — | — | — | — | — | |
| VGG-Face descriptorsupervision=Supervised2018.08 | 11.65 | 18.35 | 8.36 | 8.24 | — | — | — | — | — | — | |
| Random Forest2017.02 | 12.26 | — | — | — | — | — | — | — | — | — | |
| AVMDescription=AVM2016.11 | — | — | — | — | — | 16.75 | — | — | 60.75 | — | |
| BFGS-LDLDescription=BFGS-LDL (KL)2016.11 | — | — | — | — | 7.21 | 8.72 | 12.69 | 90.62 | 86.81 | 79.8 | |
| C-ConvNetDescription=C-ConvNet (softmax)2016.11 | — | — | — | — | 7.87 | 9.34 | 13.65 | 87.75 | 83.79 | 75.04 | |
| ConvNet+LDDescription=ConvNet+LD (a-div)2016.11 | — | — | — | — | 6.55 | 7.02 | 10.77 | 92.8 | 91.88 | 86.14 | |
| ConvNet+LSDescription=ConvNet+LS (KL)2016.11 | — | — | — | — | 7.69 | 9.1 | 13.33 | 88.34 | 85 | 76.47 | |
| DLDLDescription=DLDL (KL)2016.11 | — | — | — | — | 5.75 | 6.6 | 9.78 | 95.41 | 92.89 | 89.27 | |
| R-ConvNetDescription=R-ConvNet (l2)2016.11 | — | — | — | — | 6.57 | 8.44 | 11.88 | 92.84 | 84.76 | 79.56 | |
| R-ConvNetDescription=R-ConvNet (l1)2016.11 | — | — | — | — | 6.01 | 7.07 | 10.34 | 94.6 | 89.62 | 85.45 | |
| R-ConvNetDescription=R-ConvNet (e-ins)2016.11 | — | — | — | — | 5.96 | 7.13 | 10.35 | 94.94 | 90 | 86.21 |