350 rub
Journal Neurocomputers №8 for 2013 г.
Article in number:
Superresolution on mobile devices using remote GPU clusters
Keywords: superresolution GPU image enhancing belief propagation Мarkov random filed
Authors:
S.N. Zagoruyko, P.V. Skribtsov, A.V. Dolgopolov
Abstract:
Image enhancing techniques such as superresolution or deblurring would be in common use on mobile devices although because of limited performance of mobile devices it is not used. In the meantime, GPU devices and GPU clusters are highly suitable for image enhancing and recent approaches show about 10x and more higher performance in this problems. So we propose the algorithm that gives an ability to use powerful GPU clusters from mobile devices for example-based superresolution. Proposed algorithm uses a client application to send an image to the server application, which uses GPU-based searching and Markov Random Fields with Belief Propagation optimization to upscale the image and send it back to the client application.
Pages: 42-49
References

  1. Keys R. Cubic convolution interpolation for digital image processing // IEEE Trans. Acoustics, Speech, SignalProcessing. 1981. V. 29. № 6. R. 1153-1160.
  2. Li X. and Orchard M. T. New edge-directed interpolation // IEEE Trans. Image Processing. 2001. V. 10. № 10. R. 1521-1527.
  3. Freeman W. T., Jones T. R., and Pasztor E. C. Example-based super-resolution // IEEE Computer Graphics and Applications. 2002. V. 22. № 2. R. 56(65.
  4. Baker S. and Kanade T. Limits on super-resolution and how to break them // IEEE Trans. Pattern Analysis and Machine Intelligence. 2002. V. 24. № 9. R. 1167-1183
  5. Freeman W. T., Pasztor E. C., and Carmichael O. T. Learning low-level vision // International Journal of Computer Vision. 2000. V. 40. № 1. R. 25-47.
  6. Baker S. and Kanade T. Limits on super-resolution and how to break them // IEEE Trans. Pattern Analysis and Machine Intelligence. 2002. V. 24. № 9. R. 1167-1183.
  7. Hertzmann, Jacobs C. E., Oliver N., Curless B., and Salesin D. H. Image analogies // In Computer Graphics (Proc. Siggraph 2001). NY.: ACM Press. 2001. R. 327-340
  8. Pickup L. C., Roberts S. J., and Zissermann A. A sampled texture prior for image super-resolution // I, Advances in Neural Information Processing Systems (S. Thrun, L. Saul, and B. Scholkopf, eds). Cambridge. MA. 2004. MIT Press.
  9. .              Kim K. I., Kim D. H., and Kim J.-H. Example-based learning for image super-resolution // In Proc. the third Tsinghua-KAIST Joint Workshop on Pattern Recognition. 2004. R. 140-148.
  10. K. Ni and T. Q. Nguyen Image superresolution using support vector regression // IEEE Trans. Image Processing. 2007. V. 16. № 6. R. 1596-1610.
  11. Kim K. I. and Kwon Y.  Example-based learning for single image super-resolution // In Proc. DAGM. 2008. R. 456-465
  12. Kim K. I. and Kwon Y.  Example-based Learning for Single-Image Super-Resolution and JPEG Artifact Removal.
IEEE Trans. Pattern Analysis and Machine Intelligence. 2010. V. 32. № 6. R. 1127(1133,
  13. Zagorujko S. N. Uskorenie vy'chisleniya by'strogo preobrazovaniya Fur'e na graficheskix proczessorax // 54-ya nauchnaya konf. MFTI. 2011.
  14. Brunton A., Shu C., and Roth G. Belief propagation on the GPU for stereo vision // In Proc. 3rd Canadian Conf. Computer and Robot Vision. 2006. R. 76.
  15. Grauer-Gray S., Kambhamettu C. and Palaniappan K. GPU implementation of belief propagation using CUDA for cloud tracking and reconstruction // In Proc. PRRS2008. 2008. R. 1-4.
  16. Govindaraju N. K. GPUFFTW: High performance GPU-based FFT library // In Supercomputing. 2006
  17. Nene S. A. and Nayar S. K. A Simple Algorithm for Nearest Neighbor Search in High Dimensions // IEEE Trans. Pattern Analysis and Machine Intelligence. Sept. 1997. V. 19. № 9. R. 989(1003.
  18. E'lektronny'j resurs http://opencv.willowgarage.com
  19. E'lektronny'j resurs http://www.nvidia.com/content/cuda/cuda-toolkit.html