Title: | A GPU-accelerated augmented Lagrangian based L1-mean curvature Image denoising algorithm implementation |
Authors: | Myllykoski, Mirko Glowinski, Roland Kärkkäinen, Tommi Rossi, Tuomo |
Citation: | WSCG 2015: full papers proceedings: 23rd International Conference in Central Europeon Computer Graphics, Visualization and Computer Visionin co-operation with EUROGRAPHICS Association, p. 119-128. |
Issue Date: | 2015 |
Publisher: | Václav Skala - UNION Agency |
Document type: | konferenční příspěvek conferenceObject |
URI: | wscg.zcu.cz/WSCG2015/CSRN-2501.pdf http://hdl.handle.net/11025/29433 |
ISBN: | 978-80-86943-65-7 (print) 978-80-86943-61-9 (CD-ROM) |
ISSN: | 2464–4617 (print) 2464–4625 (CD-ROM) |
Keywords: | rozšířená Lagrangianova metoda;GPU výpočty;odstranění šumu z obrazu;zpracování obrazu;střední zakřivení;OpenCL |
Keywords in different language: | augmented Lagrangian method;GPU computing;image denoising;image processing;mean curvature;OpenCL |
Abstract: | This paper presents a graphics processing unit (GPU) implementation of a recently published augmented Lagrangian based L1-mean curvature image denoising algorithm. The algorithm uses a particular alternating direction method of multipliers to reduce the related saddle-point problem to an iterative sequence of four simpler minimization problems. Two of these subproblems do not contain the derivatives of the unknown variables and can therefore be solved point-wise without inter-process communication. In particular, this facilitates the efficient solution of the subproblem that deals with the non-convex term in the original objective function by modern GPUs. The two remaining subproblems are solved using the conjugate gradient method and a partial solution variant of the cyclic reduction method, both of which can be implemented relatively efficiently on GPUs. The numerical results indicate up to 33-fold speedups when compared against a single-threaded CPU implementation. The pointwise treated subproblem that takes care of the non-convex term in the original objective function was solved up to 76 times faster. |
Rights: | © Václav Skala - UNION Agency |
Appears in Collections: | WSCG 2015: Full Papers Proceedings |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Myllykoski.pdf | Plný text | 3,16 MB | Adobe PDF | View/Open |
Please use this identifier to cite or link to this item:
http://hdl.handle.net/11025/29433
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.