Can we unify monocular detectors for autonomous driving by using the pixel-wise semantic segmentation of CNNs?
Eduardo RomeraLuis Miguel BergasaRoberto ArroyoPublished in: CoRR (2016)
Keyphrases
- pixel wise
- superpixels
- pixel level
- object detection
- foreground objects
- object segmentation
- image sequences
- multiscale
- spatial information
- input image
- object categories
- background subtraction
- natural images
- long range
- segmentation algorithm
- segmentation method
- appearance model
- conditional random fields
- bounding box
- optical flow
- image analysis
- template matching
- multiple images
- higher order
- image patches
- graph structure
- multiple objects