Lifting AutoEncoders: Unsupervised Learning of a Fully-Disentangled 3D Morphable Model using Deep Non-Rigid Structure from Motion.
Mihir SahasrabudheZhixin ShuEdward BartrumRiza Alp GülerDimitris SamarasIasonas KokkinosPublished in: CoRR (2019)
Keyphrases
- non rigid structure from motion
- unsupervised learning
- d morphable model
- active appearance models
- shape model
- denoising
- semi supervised
- facial images
- low resolution
- human faces
- appearance model
- dimensionality reduction
- deformable models
- face recognition
- statistical model
- single image
- feature selection
- motion analysis
- graph cuts
- parametric models
- least squares
- image segmentation
- image processing