Abstract
Catadioptric omnidirectional view sensors have found increasing adoption in various robotic and surveillance applications due to their 360° field of view. However, the inherent distortion caused by the sensors prevents their direct utilisations using existing image processing techniques developed for perspective images. Therefore, a correction processing known as “unwrapping” is commonly performed. However, the unwrapping process incurs additional computational loads on central processing units. In this paper, a method to reduce this burden in the computation is investigated by exploiting the parallelism of graphical processing units (GPUs) based on the Compute Unified Device Architecture (CUDA). More specifically, we first introduce a general approach of parallelisation to the said process. Then, a series of adaptations to the CUDA platform is proposed to enable an optimised usage of the hardware platform. Finally, the performances of the unwrapping function were evaluated on a high-end and low-end GPU to demonstrate the effectiveness of the parallelisation approach.
Similar content being viewed by others
Notes
The coalesced access of the unwrapping kernel can be confirmed using the NVIDIA visual profiler [17].
References
Baker, S., Nayar, S.K.: A theory of single-viewpoint catadioptric image formation. Int. J. Comput. Vis. 35(2), 175–196 (1999)
Bui, P., Brockman, J.: Performance analysis of accelerated image registration using gpgpu. In: Proceedings of 2nd Workshop on General Purpose Processing on Graphics Processing Units, ACM, New York, GPGPU-2, pp. 38–45 (2009). doi:10.1145/1513895.1513900
Burden, R.L., Faires, D.J.: Numerical Analysis, 7th edn. Brooks Cole, Belmont (2000)
Chen, L.D., Zhang, M.J., Xiong, Z.H.: Series-parallel pipeline architecture for high-resolution catadioptric panoramic unwrapping. IET Image Process. 4(5):403–412 (2010). doi:10.1049/iet-ipr.2009.0286
Chong, N.S., Kho, Y.H., Wong, M.L.D.: A closed form unwrapping method for a spherical omnidirectional view sensor. EURASIP J. Image Video Process. 2013, 5 (2013). doi:10.1186/1687-5281-2013-5, url:http://jivp.eurasipjournals.com/content/2013/1/5
Chong, N.S., Kho, Y.H., Wong, M.L.D.: Custom aspect ratio correction for unwrapped omnidirectional view images. Comput. Electr. Eng. (2013, in press). doi:10.1016/j.compeleceng.2013.04.005, url:http://dx.doi.org/10.1016/j.compeleceng.2013.04.005
Gaspar, J., Santos-Victor, J.: Visual path following with a catadioptric panoramic camera. In: Proceedings of the International Symposium on Intelligent Robotic Systems—SIRS’99, Coimbra (1999). url:http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.33.3379
Hartley, R.I., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press, Cambridge (2004)
Hicks, R.A., Bajcsy, R.: Reflective surfaces as computational sensors. Image Vis. Comput. 19(11), 773–777 (2001). url:http://linkinghub.elsevier.com/retrieve/pii/S0262885600001049
Intel Corporation: Intel threading building blocks. Version 4.0 (2012)
Jeng, S.W., Tsai, W.H.: Using pano-mapping tables for unwarping of omni-images into panoramic and perspective-view images. Image Process. IET 1(2), 149–155 (2007). doi:10.1049/iet-ipr:20060201
Lei, J., Du, X., Zhu, Y.F., Liu, J.L.: Unwrapping and stereo rectification for omnidirectional images. J. Zhejiang Univ. Sci. A 10(8), 1125–1139 (2009). url:http://www.springerlink.com/index/10.1631/jzus.A0820357
Li, J., Lu, Y., Pu, B., Xie, Y., Qin, J., Pang, W.M., Heng, P.A.: Accelerating active shape model using gpu for facial extraction in video. In: Proceedings of the IEEE International Conference on Intelligent Computing and Intelligent Systems (ICIS 2009), vol. 4, pp. 522–526 (2009). doi:10.1109/ICICISYS.2009.5357636
Nayar, S.K.: Catadioptric omnidirectional camera. In: Proceedings of the 1997 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 482–488 (1997). doi:10.1109/CVPR.1997.609369
NVIDIA Corporation: CUDA GPU occupancy calculator. Version 3.0 (2012a)
NVIDIA Corporation: NVIDIA CUDA C Programming Guide. NVIDIA Corporation, Santa Clara (2012b)
NVIDIA Corporation: NVIDIA visual profiler. Version 4.2 (2012c)
Rossi, R., Savatier, X., Ertaud, J.Y., Mazari, B.: Real-time 3d reconstruction for mobile robot using catadioptric cameras. In: Proceedings of the IEEE International Workshop on Robotic and Sensors Environments (ROSE 2009), pp. 104–109 (2009). doi:10.1109/ROSE.2009.5355981
Scaramuzza, D., Martinelli, A., Siegwart, R.: A toolbox for easily calibrating omnidirectional cameras. In: Proceedings of the 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems, Beijing, pp. 5695–5701 (2006). doi:10.1109/IROS.2006.282372
Scherl, H., Keck, B., Kowarschik, M., Hornegger, J.: Fast gpu-based ct reconstruction using the common unified device architecture (cuda). In: Proceedings of the IEEE Nuclear Science Symposium Conference Record (NSS ’07), vol. 6, pp. 4464–4466 (2007). doi:10.1109/NSSMIC.2007.4437102
Zhang, H., Xie, Y., Heng, P.A.: Accelerating feature extraction for patch-based multi-view stereo algorithm. In: Proceedings of the International Conference on Computer Design and Applications (ICCDA), vol. 5, pp. 511–515 (2010). doi:10.1109/ICCDA.2010.5541068
Zhang, Z.: A flexible new technique for camera calibration. IEEE Trans. Pattern Anal. Mach. Intell. 22(11), 1330–1334 (2000). doi:10.1109/34.888718
Acknowledgments
Nguan Soon Chong thanks Swinburne University of Technology (Sarawak Campus) for his Ph.D. studentship. The authors would like to personally thank Jian Soon Ng and Albin Sui Hian Kuek for their useful discussion on this research topic.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chong, N.S., Wong, M.L.D. & Kho, Y.H. Accelerated catadioptric omnidirectional view image unwrapping processing using GPU parallelisation. J Real-Time Image Proc 12, 55–69 (2016). https://doi.org/10.1007/s11554-013-0390-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11554-013-0390-x