Charles River Analytics received funding from the Air Force Research Lab to develop the Stereo Anaglyph Creation via Image Sequences to Image Pairs (ISIP) project. The Air Force requires a system that automatically analyzes image sequences from a ground-based telescope or a satellite camera and selects pairs of images of uncooperative space objects that pass overhead. ISIP provides a solution to this need by analyzing the image sequence in order to determine the space object’s shape and motion. It then selects the pair of images that produces the best stereo image. As an additional benefit, ISIP can build a 3D model from the chosen image pair using a dense stereo reconstruction technique. Charles River Analytics is developing ISIP in C++ as a near-real-time capable fully automatic system.
Charles River Analytics is uniquely qualified for this project with its extensive experience in computer vision, in particular structure from motion, stereo processing, and target tracking for numerous imaging modalities.