Improving Hough Based Pedestrian Detection Accuracy by Using Segmentation and Pose Subspaces.
Jarich VansteenbergeMasayuki MukunokiMichihiko MinohPublished in: IEICE Trans. Inf. Syst. (2014)
Keyphrases
- detection accuracy
- face detection
- detection algorithm
- processing speed
- object detection
- hough voting
- number of false alarms
- human pose tracking
- detection rate
- segmentation algorithm
- parameter space
- multiscale
- false detections
- level set
- hough transform
- object segmentation
- pose estimation
- segmentation method
- image segmentation
- pedestrian detection
- shape prior
- edge detection
- high dimensional data
- face detector
- computer vision
- data mining
- medical images
- multi class
- d objects
- feature extraction
- low rank representation
- feature selection