Context-constrained matching of hierarchical CAD-based models for outdoor scene interpretation

The authors propose a method for outdoor scene interpretation which recognizes many kinds of objects in the scene, and determines their geometric structures using context information of objects in the scene and flexible matching of hierarchical CAD-based models. The system can deal with outdoor scenes with artificial objects which may have various shapes and sizes, and natural objects for which there are no CAD-based models. The system uses the intermediate results of the scene interpretation as context information to prune the candidate object classes, their positions and orientations, and uses hierarchical CAD models so that higher models may be able to represent one object class such as a house, with several variations by allowing changes of dimensions. The input scene is segmented into a ground plane and objects on it using the multisensory information such as video and range images, and planar surface patches are extracted from the segmented regions. Flexible matching between the scene descriptions and the hierarchical CAD models is performed