Top-Down Shape Abstraction Based on Greedy Pole Selection.
Zhiyang DouShiqing XinRui XuJian XuYuanfeng ZhouShuang-Min ChenWenping WangXiuyang ZhaoChanghe TuPublished in: CoRR (2019)
Keyphrases
- high level
- forward selection
- greedy algorithm
- shape matching
- greedy algorithms
- data sets
- shape model
- dynamic programming
- prior information
- feature selection
- shape analysis
- shape features
- shape representation
- shape change
- data abstraction
- shape recognition
- locally optimal
- neural network
- search space
- figure ground segmentation
- genetic algorithm
- shape retrieval
- perceptual grouping
- medial axis
- computer vision
- image segmentation