The Optimization of Interface Interactivity using Gesture Prediction Engine

The primary objective of this project is to develop a gesture recognition engine for interactive interfaces using Microsoft Kinect device. A photo album is a sample of daily-use applications that is capable of having interactive interface. In this project there are features implemented to help users to view and edit their photos on the easier way. Although the 3D interface of photo album increases the reality and easy to use, simplicity of natural gestures which are recognizing by the gesture recognition engine eases the interaction. The contribution of this project is simultaneous work of a prediction and recognition engine. The algorithm benefits a Hidden Markov Model (HMM) state machine to record, update and calculate the occurrence probability of each gesture as a state in relation with previous states. It also aims to solve a major problem of interaction with the same applications which were their dependence on using devices physically and touch them directly. The optimized model had tested in an interactive digital space.