filename : Ger07.pdf entry : inproceedings conference : 3DIM 2007, Montréal, Canada, 21-23 August, 2007 pages : 8 year : 2007 month : August title : Automatic Pose Estimation for Range Images on the GPU subtitle : author : Marcel Germann, Michael D. Breitenstein, In Kyu Park, Hanspeter Pfister booktitle : 3DIM '07: Proceedings of the Sixth International Conference on 3-D Digital Imaging and Modeling ISSN/ISBN : 0-7695-2939-4 editor : Guy Godin, Patrick Hébert, Takeshi Masuda, Gabriel Taubin publisher : IEEE Computer Society publ.place : Washington, DC, USA volume : issue : language : English keywords : pose estimation, GPU, depth images, range images abstract : Object pose (location and orientation) estimation is a common task in many computer vision applications. Although many methods exist, most algorithms need manual initialization and lack robustness to illumination variation, appearance change, and partial occlusions. We propose a fast method for automatic pose estimation without manual initialization based on shape matching of a 3D model to a range image of the scene. We developed a new error function to compare the input range image to pre-computed range maps of the 3D model. We use the tremendous dataparallel processing performance of modern graphics hardware to evaluate and minimize the error function on many range images in parallel. Our algorithm is simple and accurately estimates the pose of partially occluded objects in cluttered scenes in about one second.