Infrared Pedestrian Segmentation Through Background Likelihood and Object-Biased Saliency.
Lu LiFugen ZhouXiangzhi BaiPublished in: IEEE Trans. Intell. Transp. Syst. (2018)
Keyphrases
- infrared
- infrared images
- background pixels
- foreground and background
- object segmentation
- infrared imagery
- foreground objects
- bounding box
- figure ground
- multi sensor
- pedestrian tracking
- multiple objects
- object contours
- pedestrian detection and tracking
- salient objects
- visible spectrum
- image regions
- object detection
- segmentation algorithm
- pixel level
- electro optical
- thermal infrared
- background subtraction
- target detection and tracking
- image segmentation
- image analysis
- segmentation method
- salient regions
- d objects
- hyperspectral
- infrared camera
- saliency map
- multiscale
- edge detection
- moving objects
- visual attention
- graph cuts
- lighting conditions
- image fusion