Using GPUs to accelerate RGB-D mapping

• Autorzy: Majek K. , Będkowski J.
• Języki publikacji: EN

Abstrakty

EN

Since the low-cost depth sensors have become widely available, RGB-D mapping is getting more popular in mobile robotics. These sensors provide large amounts of data providing images and depth information. To build 3D maps on-line the high speed data registration alogrithm is needed. To solve this problem we propose the GPGPU implementation of the FAST and BFROST matching algorithms using both RGB image and normal vectors visualization image. We compute the initial transformation usisng RANSAC and HORN closed-form solution algorithm. Finally, we use semantic information to improve the GPGPU ICP and to get an accurate model of the environment.

Słowa kluczowe

EN

image perception; RGB-D; GPU

PL

percepcja obrazu; obraz RGB-D; GPU

• Wydawca: Oficyna Wydawnicza Politechniki Warszawskiej
• Czasopismo: Prace Naukowe Politechniki Warszawskiej. Elektronika
• Rocznik: 2014
• Tom: z. 194, t. 1
• Strony: 307--316
• Opis fizyczny: Bibliogr. 14 poz., rys., tab.

Twórcy

autor

Majek K.

• Institute of Mathematical Machines, Ludwika Krzywickiego 34, 02-078 Warsaw, Poland

autor

Będkowski J.

• Institute of Mathematical Machines, Ludwika Krzywickiego 34, 02-078 Warsaw, Poland

Bibliografia

• [1] http://www.3d-scanner.net, Fraunhofer IAIS, 2014.
• [2] Automation Group (Jacobs University Bremen) and Knowledge-Based Systems Group (University of Osnabruck), 3DTK - The 3D Toolkit, http://slam6d.sourceforge.net/, February 2011.
• [3] H. Bay, T. Tuytelaars, and L. V. Gool. SURF: Speeded up robust features, ECCV, 2006, 404-417.
• [4] P. J. Besl, H. D. Mckay. A method for registration of 3-d shapes. In Pattern Analysis and Machine Intelligence, IEEE Transactions on, 14 (2) (1992), 239-256.
• [5] Będkowski, J., Majek, K., Masłowski, A. and Kaczmarek, P. Recognition Of 3d Objects For Walking Robot Equipped With Multisense-SL Sensor Head, CLAWAR, 2013.
• [6] Będkowski, J., Majek, K., and Nuchter, A. A. General Purpose Computing on Graphics Processing Units for Robotic Applications. In Journal of Software Engineering for Robotics, Vol 4, No 1 (2013).
• [7] Będkowski J., Maslowski A. and De Cubber G. Real time 3D localization and mapping for USAR robotic application. In Industrial Robot: An International Journal, 5 (2012), 464-474, 2012.
• [8] Cronje, J. BFROST: binary features from robust orientation segment tests accelerated on the GPU. 2011.
• [9] Fischler, M. A., and Bolles, R. C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography In Communications of the ACM, 24(6), 381-395, 1981.
• [10] Horn, B. K. Closed-form solution of absolute orientation using unit quaternions. JOSA A, 4(4), 629-642, 1 987.
• [11] S. Leutenegger, M. Chli, and R. Siegwart, BRISK: Binary robust invariant scalable keypoints In Computer Vision (ICCV)2011 IEEE International Conference.
• [12] http://pointclouds.org/, PCL, Point Cloud Library, 2014.
• [13] E. Rosten and T. Drummond. Fusing points and lines for high performance tracking. In IEEE International Conference on Computer Vision, vol. 2, 1508-1511, October 2005.
• [14] http://velodynelidar.com/lidar/lidar.aspx, Velodyne, 2014.