Object Detection in Indoor Scenes Using Log-Polar Mapping

In this paper a system that successfully detects stationary objects in indoor scenes is presented. The doors and ceiling lamps have been chosen to be detected in the initial phase of our system. The scene was divided into four diierent regions. The edges resulting from the walls/ceiling and the walls//oor intersections are the boundaries that separate the diierent regions. The use of the log-polar mapping transforms the problem of nding four edges of unknown orientations, that separated the diierent regions, into nding four horizontal edges. The log-polar map was performed such that it possesses the varying resolution property. The objects were detected using model-matching in the log-polar domain. The varying resolution property of the mapping simpliied the model-matching problem. The detection (of edges and then objects) was very fast due to the small size of the log-polar image (64 64) pix-els.