On the power of data augmentation for head pose estimation

Benchmark Model Rank Results
face-alignment-on-aflw2000-3dOpNet#8Balanced NME (2D Sparse Alignment): 3.55%
head-pose-estimation-on-aflw2000OpNet#1MAE: 3.15Geodesic Error (GE): 5.23
head-pose-estimation-on-biwiOpNet#8MAE (trained with other data): 3.57Geodesic Error (GE): 7.01