Fixation and Saccade Based Face Recognition from Single Image per Person with Various Occlusions and Expressions.
Xingjie WeiChang-Tsun LiPublished in: CVPR Workshops (2013)
Keyphrases
- single image
- eye movements
- face recognition
- facial expressions
- partially occluded
- visual search
- pose variations
- eye tracking
- computer vision
- visual attention
- face images
- multiple images
- d scene
- shape from shading
- human faces
- shape recovery
- super resolution
- light source
- image pairs
- d morphable model
- feature extraction
- principal component analysis
- image sequences
- real scenes
- facial images
- face detection
- partial occlusion
- structured light
- intrinsic images
- keypoints
- stereo images
- saliency map
- d objects
- optical flow
- high quality
- cast shadows