作者: Ali Azarbayejani , Tinsley Galyean , Alex Pentland
DOI:
关键词: Tracking (particle physics) 、 Series (mathematics) 、 Computer vision 、 Focal length 、 Artificial intelligence 、 Translation (geometry) 、 Rotation (mathematics) 、 Object (computer science) 、 Geography 、 Reference frame 、 Segmentation
摘要: A method and apparatus for generating three-dimensional, textured computer models from a series of video images an object is disclosed. The invention operates by tracking selected group features through image frames and, based on changes in their relative positions, estimates parameters specifying camera focal length, translation rotation, the positions tracked reference frame. After segmentation into two-dimensional bounded regions that each correspond to discrete surface component actual object, texture contained various applied these produce final three-dimensional model both geometrically photometrically specified.