SelFSR: Self-Conditioned Face Super-Resolution in the Wild via Flow Field Degradation Network.
Xianfang ZengJiangning ZhangLiang LiuGuangzhong TianYong LiuPublished in: CoRR (2021)
Keyphrases
- super resolution
- flow field
- low resolution
- high resolution
- optic flow
- motion estimation
- multi frame
- optical flow
- image restoration
- image reconstruction
- super resolution reconstruction
- high resolution images
- single image
- low resolution images
- degraded images
- optical flow estimation
- motion blur
- image sequences
- motion analysis
- high quality
- face images
- motion segmentation
- depth map
- image super resolution
- light field
- motion field
- human faces
- computer vision
- motion parameters
- face recognition
- facial expressions
- feature selection
- image quality