Human-Body Gestures Recognition has been a challenge for several years. Gesture recognition is one of the dimensions of the Digital Image Processing. The proposed project aims to sense the position of the human body parts, e.g., Head, Shoulder, Hands, Joints, Legs, etc. The position of the skeletal parts of the human body in front of the camera of the smartphone, are detected possibly using Android compatible OpenCV or some other free open-source library.
Although Kinect has advantage over Smart phone’s camera because it has got capability of automatically detecting (built-in feature) the depth map using color gradients from white (near) to blue (far). As smart phone. However smart phone’s big benefit is that it is available in almost every person’s pocket. Therefore, using smart phone’s camera for gesture detection can be used for many applications (games, physical exercises, device controlling, etc.). In order to work like Kinect Smart phone’s camera has to have good quality lens and good resolution, however detecting an object at the distance of two to three meters with the help of AI recognition techniques + Heuristics, is quite possible and doable. This could be a step toward replacing Kinect in certain applications. Something as shown in the following Figure, however recognition should be done through Android Phone’s camera instead of Kinect camera.
Emoticon Emoticon
Note: only a member of this blog may post a comment.