Automated System for Semantic Object Labeling With Soft-Object Recognition and Dynamic Programming Segmentation.
Jonas ClevelandDinesh ThakurPhilip M. DamesCody J. PhillipsTerry KientzKostas DaniilidisJohn BergstromVijay KumarPublished in: IEEE Trans Autom. Sci. Eng. (2017)
Keyphrases
- scene understanding
- object recognition
- object detection
- object segmentation
- dynamic programming
- d objects
- image segmentation
- multiple objects
- object categories
- man made objects
- object description
- image labeling
- computer vision
- object representation
- coarse to fine
- object matching
- semantic segmentation
- natural images
- graph cut segmentation
- object class
- segmentation algorithm
- keypoints
- object model
- multiscale
- signed distance
- medical images
- bounding box
- pixel level
- fully automatic
- interactive image segmentation
- figure ground segmentation
- automated segmentation
- level set
- image regions
- labeling process
- scene classification
- low level features
- image segmentations
- region growing
- object classes
- recognizing objects
- image segments
- rib cage
- active learning
- global shape
- object contours
- object shape
- target object
- natural language
- shape prior
- object tracking
- segmentation method
- automated analysis
- foreground and background
- moving objects