作者: Jacob Feldman , Manish Singh , Erica Briscoe , Vicky Froyen , Seha Kim
DOI: 10.1007/978-1-4471-5195-1_4
关键词:
摘要: We present a unified Bayesian approach to shape representation and related problems in perceptual organization, including part decomposition, similarity, figure/ground estimation, 3D shape. The is based on the idea of estimating skeletal structure most likely have generated observed via process stochastic “growth.” survey briefly show how it can be extended principled way solve wide array problems.