Happiness understanding system based on human action recognition

The daily human action understanding system based on HAPPINESS factors is presented for happiness promotion using human action recognition. Recognition and understanding of human behavior is a popular research topic in computer vision. The proposed system extracts the features of skeleton information using the entire body and arm movement. Dynamic Time Warping is applied to compare the similarity between the two actions in action recognition. There are two contributions in this work, 1) A suitable templates selection among different subjects regarding as a representative human HAPPINESS action. And Non-Static Sequence Segmentation is proposed for action recognition. The system classifies the corresponding HAPPINESS factors from the results of action recognition. Experiments were performed on online test and the results show that the accuracy is 84.81%.