研究实例：非刚性配准.pptx

研究实例：非刚性配准;定义;特性;研究的缘由;分类;非迭代法;迭代法;迭代法;非迭代法;Shape Context Method;Shape Context Method;Shape Context Method;Shape Context Method;Shape Context Method;Shape Context Method;Shape Context Method;Shape Context Method;Shape Context Method：Limitation;Feature in Shape Context Method;类似方法：Graph Based Method;迭代法;迭代法的优点和缺点;思考：上述Shape context method 和 Graph based methods是否可以转化为迭代法？;Iterative Closest Point (ICP)：定义;Iterative Closest Point (ICP)：定义;Iterative Closest Point (ICP);如何评估算法（基础评价）;Iterative Closest Point (ICP): Limitation;TPS-RPM（Robust Point Matching）Chui et al. 2003;TPS-RPM: Fuzzy Correspondence;TPS-RPM: Annealing（退火法）;TPS-RPM: Annealing（退火法）;TPS-RPM: 对应评估;TPS-RPM: 更新几何变换;TPS-RPM: 流程;TPS-RPM: Demo;Performance Estimation;TPS-RPM vs. ICP;TPS-RPM vs. Shape Context Method;休息10分钟;CPD（Coherent Point Drift）Myronenko et al. 2010;CPD（Coherent Point Drift）：一致性;CPD（Coherent Point Drift）：EM算法;CPD（Coherent Point Drift）：EM算法;CPD（Coherent Point Drift）：算法流程;CPD（Coherent Point Drift）：Limitation;CPD vs. TPS-RPM (Fish);GMMREG（Gaussian Mixture Model Registration）Jian et al. 2011;GMMREG: Gaussian Mixture Model Registration;GMMREG: Gaussian Mixture Model Registration;GMMREG: Performance;GMMREG: Limitation;Current Methods：Limitation;Features;A robust global and local mixture distance based non-rigid point set registration;Global Distance;Local Distance;Mixture Distance;Correspondence Estimation;Transformation Updating;Annealing （退火）;算法流程（GLMDTPS）;方法对比;实验 1-Four Point Sets;实验 1-Wide Range;实验 1-Partial Matching;实验 1-With Different Parameter K：High Sampling Rate;实验 1-With Different Parameter K：Noise，Outlier and Rotation;实验 2-Sequence Images： CMU Datasets;实验 3-Real Images;实验 4-计算复杂度;实验 4-退火系数;实验 5-计算时间;GLMDTPS：Demo