作者: Maneesh Agrawala , Richard Szeliski , David H. Salesin , Aseem O. Agarwala , Michael F. Cohen
DOI:
关键词: Artificial intelligence 、 Computer vision 、 Architecture 、 Surface (mathematics) 、 Computer science 、 Image (mathematics) 、 Computer graphics (images) 、 Transition (fiction) 、 Subject matter 、 Single image
摘要: The claimed subject matter can provide an architecture that facilitates producing a single image visualize scene too large to depict from any perspective view. These images be stitched together on 2-dimensional picture surface form multi-perspective of the entire extent depicted by input images. Where respective overlap, objective function employed determine where make transition between one and next. employ data cost seam this determination.