Learning and matching of dynamic shapes manifolds for human action recognition project explain about introducing a new concept for detecting a human action more accurately compare to previous methods.
This application is most useful in sports, surveillance, perceptual interfaces.
Using this system we can detect track and recognize people and more importantly it is easy to understand human activities by analyzing from image sequences.
In previous system there are two categories where higher level analysis is carried out. Template matching based approaches and state space approaches.
This application is implemented in Microsoft visual studio .net(C# .Net ) platform.
Matching of dynamic shapes manifolds for human action recognition .Net Project Reference documents.