Recognizing Personal Location from Video
Hisashi Aoki, Bernt Schiele, Alex Pentland · 1998
An important function for wearable computers is the recognition of places and locations. This paper proposes an image sequence matching technique for the recognition of previously visited places. Similar in spirit to single word recognition in speech recognition, a dynamic programming algorithm is proposed for the calculation of dissimilarities of video sequences. Such video sequences represent not only the place itself but also the approaching trajectory. The algorithm uses a simple and robust representation of a single frame without compromising the discrimination between different places. Preliminary experimental results demonstrate the discrimination and the robustness of the approach with respect to the angle of the approaching trajectory. 1. Introduction Computers and cameras are becoming small enough to wear. Computers are being equipped with cameras, microphones and other sensors, which can be used to interpret what people are doing, where they are and with whom they are inte...