Improved Isomap Algorithm for Motion Analysis

摘要

Euclidean distance, Hausdorff distance and SSP distance are discussed, and SSP distance is used to improve Isomap algorithm. Two methods are put forward for improving Isomap algorithm. One is aligning input data of original Isomap algorithm, the other is modifying Isomap algorithm itself. SSP distance is used to search neighbors and compose neighborhood graph, and the plot for each dimension of Isomap representation is used for visualization of Isomap representation. Motion analysis experiments results show that improved Isomap algorithm is better than original Isomap algorithm for translated data and has better visualization results of Isomap representation.