FCSR-GAN: Joint Face Completion and Super-resolution via Multi-task Learning.
Jiancheng CaiHu HanShiguang ShanXilin ChenPublished in: CoRR (2019)
Keyphrases
- super resolution
- multi task learning
- multi task
- low resolution
- high resolution
- image reconstruction
- learning tasks
- gaussian processes
- face images
- low resolution images
- learning problems
- motion estimation
- transfer learning
- high quality
- learning algorithm
- high order
- depth map
- learning models
- theoretical analysis
- facial expressions
- video sequences
- image patches
- facial images
- inductive learning
- data sets
- closed form
- multi class
- reinforcement learning
- three dimensional
- computer vision