Geometry feature generation based on unsupervisely detected landmarks. [1]
Disentangle bottleneck features into category-invariant features and category-specific features. Category-invariant features encode the pose information.
Reference
- Wayne Wu, Kaidi Cao, Cheng Li, Chen Qian, Chen Change Loy: TransGaGa: Geometry-Aware Unsupervised Image-To-Image Translation. CVPR 2019