Non-rigid Point Set Registration via Coherent Spatial Mapping and Local Structures Preserving
2016 15th International Symposium on Parallel and Distributed Computing (ISPDC)(2016)
摘要
Non-rigid point set registration is a fundamental problem for many computer vision technologies. In this paper, we proposed a new non-rigid point set registration method based on coherent spatial mapping (CSM) and local geometrical constraint. Our central idea is to express each point as a weighted sum of several nearest neighbors and the same relation holds after the transformation. The registration problem is solved by minimizing an error function, which combines the the global model and local geometrical constraint. The registration experiments are undertaken on various synthetic and real data. The results demonstrate that the proposed approach is robust and is superior to the state-of-the-art methods.
更多查看译文
关键词
non-rigid point set registration,coherent spatial mapping,CSM,local structures preserving,computer vision,transformation,error function minimization
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络