SUBSPACE METHODS FOR ROBOT VISION

In contrast to the traditional approach, visual recognition is formula ted as one of matching appearance rather than shape. For any given rob ot vision task, all possible appearance variations define its visual w orkspace. A set of images is obtained by coarsely sampling the workspa ce. The image set is compressed to obtain a low-dimensional subspace, called the eigenspace, in which the visual workspace is represented as a continuous appearance manifold. Given an unknown input image, the r ecognition system first projects the image to eigenspace. The paramete rs of the vision task are recognized based on the exact location of th e projection on the appearance manifold. An efficient algorithm for fi nding the closest manifold point is described, The proposed appearance representation has several applications in robot vision. As examples, a precise visual positioning system, a real-time visual tracking syst em, and a real-time temporal inspection system are described.