Basic Research in Computer Science: Image Understanding

Abstract : The CMU Image Understanding program performed basic research and technology development toward robust, flexible, and precise vision systems to impact a wide variety of military and civilian applications. The accomplished results include: hypergeometric filter-based image matching; multi-body factorization for structure from motion; a trainable face detection system; 3D surface representation from multiple range images; learning of an object appearance model: recognition of 3D objects in range images by the Spin Image method; eigen window method for SAR image recognition; and shape matching technique and its medical application.