An intelligent bulletin board system with real-time vision-based interaction using head pose estimation

A bulletin board is a place where people can read announcements, posters or leave public messages. However, traditional bulletin board certainly have several disadvantages such like without real-time interaction with users and monotonous. This paper presents an intelligent bulletin board system (iBBS), which allows a user to real-time interact with it without any additional auxiliaries. At first, a real-time front-view face detection using Harr-like features is used to decide when iBBS should wake up and become interactive with the user After system initialization, we keep on finding some feature points within the detected face area. Then we estimate the orientation of user's head via pyramidal Lucas-Kanade optical flow tracking. Compared to the traditional bulletin board system, our system has more flexibility. With the comparison to other non-vision-based input devices such like gloves or markers, our system offers a simple, useful and economical solution for the real-time interaction between the user and computer.