請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/60066完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 歐陽明(Ming Ouhyoung) | |
| dc.contributor.author | Rih-Ding Peng | en |
| dc.contributor.author | 彭日鼎 | zh_TW |
| dc.date.accessioned | 2021-06-16T09:53:55Z | - |
| dc.date.available | 2022-02-08 | |
| dc.date.copyright | 2017-02-08 | |
| dc.date.issued | 2017 | |
| dc.date.submitted | 2017-01-09 | |
| dc.identifier.citation | [1] C. Zhang, Z. Li, Y. Cheng, R. Cai, H. Chao, Y. Rui .,”Meshstereo: a global stereo model with mesh alignment regularization for view interpolation”. Computer Vision (ICCV), 2015 IEEE International Conference on (2015), pp. 2057–2065
[2] Scharstein, D., Szeliski, R., “A taxonomy and evaluation of dense two-frame stereo correspondence algorithms”. Microsoft Research Technical Report MSR-TR-2001-81, 2001. [3] K. Zhang, Y. Fang, D. Min, L. Sun, S. Y. Yan and Q. Tian, 'Cross-scale cost aggregation for stereo matching', Proc. CVPR [4] Q. Yang. “A non-local cost aggregation method for stereo matching,”. In CVPR,2012. [5] X. Mei, X. Sun, W. Dong, H. Wang, and X. Zhang. “Segment-tree based cost aggregation for stereo matching,”. In CVPR, 2013. [6] C. Tomasi and R. Manduchi, 'Bilateral Filtering for Gray and Color Images,' Proc.Sixth Int',l Conf. Computer Vision, pp. 839-846, Jan. 1998. [7] K.-J. Yoon and I. S. Kweon. “Adaptive support-weight approach for correspondence search,”. TPAMI, 2006. [8] K. He, J. Sun, and X. Tang. “Guided image filtering,” In ECCV, pages 1–14. 2010. [9] C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz. “Fast cost-volume filtering for visual correspondence and beyond,”. In CVPR, 2011. [10] Yang, Q. “Recursive bilateral filtering,”. ECCV 2012, pp. 399–413. Springer,Heidelberg (2012) [11] G. Saygili, L. Van Der Maaten and E. A. Hendriks, 'Stereo Similarity Metric Fusion Using Stereo Confidence', Pattern Recognition (ICPR), 2014 22nd International Conference on, pp. 2161-2166 [12] Hafner, D., Demetz, O., Weickert, J.,”Why is the census transform good for robust optic flow computation?” SSVM 2013. LNCS, vol. 7893, pp. 210–221. Springer,Heidelberg (2013) [13] Y. K. Baik, J. H. Jo and K. M. Lee., 'Fast census transform-based stereo algorithm using SSE2,'.Proc. 12th Korea Jpn. Joint Workshop Frontiers Comput. Vis. (FCV),pp. 305-309 [14] J. Han , Z. Wu , L. Li and Y. Ji., “FPGA Implementation for Binocular Stereo Matching Algorithm Based on Sobel Operator,” International Journal of Database Theory and Application Vol.9, No.4 (2016), pp.221-230 [15] O. J. Arndt, D. Becker, C. Banz, and H. Blume.” Parallel Implementation of Real-Time Semi-Global Matching on Embedded Multi-Core Architectures”. Embedded Computer Systems, pages 56-63. IEEE, 2013. [16] http://www.gnu.org/software/libc/manual/html_node/Infinity-and-NaN.html [17] http://developer.amd.com/tools-and-sdks/opencl-zone/ [18] http://vision.middlebury.edu/stereo/ [19] http://developer.amd.com/community/blog/2014/11/17/opencl-2-0-deviceenqueue/ [20] http://developer.amd.com/community/blog/2014/10/31/opencl-2-0-pipes/ | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/60066 | - |
| dc.description.abstract | 本篇論文運用 OpenCL - 平行化程式的語言來達到通用圖形處理器的目的，使運用引導影像濾波器的立體匹配演算法透過通用圖形處理器盡可能的平行加速。目前多數的立體匹配演算法依然著重在準確度，但立體匹配的應用上，對於速度也有所需求，如自動駕駛、模型掃描等。我們的演算法是局部式演算法，作為相當經典的基礎演算法，藉由這基礎的演算法來展示通用圖形處理器搭配平行化計算的加速潛力，其步驟如同經典論文 A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms 此論文所介紹的步驟，分別是匹配代價 –使用普查轉換以及索貝爾算子處理後，利用絕對誤差和計算出代價，代價聚合 -則使用引導影像濾波器來處理各個像差圖片(Disparity Image)(像差圖片空間Disparity Image Space)，最後最佳化的方式則是贏家全拿，如同一般的局部式方
法，以上步驟經過 OpenCL 的平行化後根據硬體的不同，我們整體加速至少一倍以上。這篇主要貢獻是以 OpenCL 來實作平行化運用 guided filter 的立體匹配演算法，並提供目前在新版本的平台，中央處理器、通用圖形處理器 – OpenCL 1.2 、通用圖形處理器– OpenCL 2.0 的速度比較，OpenCL 2.0 版本的特性使用解釋，以及當下最新硬體 AMD R9-390X 的紀錄。 | zh_TW |
| dc.description.abstract | In this thesis we use OpenCL which is a programming language for GPGPU to parallelize the Stereo Matching Algorithm Using Guided Filter(GF) on cost aggregation.
Today most algorithms of Stereo Matching focus on accuracy but not the processing time, there are certain applications that require real-time processing. For example - auto-driving. Our algorithm is a basic method of stereo matching which can be categorized as a local method. Its steps are similar to the thesis “A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms”. The first step is Matching-Cost. We will use Census Transform (CT), Sobel Operator (SO) and Sum of Absolute Difference (SAD). And we use Guided Filter in the second Steps: Cost Aggregation. Finally, like other local method, we apply winner-take-all. All the steps we previously describe speed up at least two times after parallelize by OpenCL. Our contribution is on the parallelization of Stereo Matching with Guided Filter and record of processing time with CPU, GPU -OpenCL 1.2 and OpenCL 2.0 on recent hardware – AMD R9 390X | en |
| dc.description.provenance | Made available in DSpace on 2021-06-16T09:53:55Z (GMT). No. of bitstreams: 1 ntu-106-R03944033-1.pdf: 1886401 bytes, checksum: acd69a7dcd9e07abbb9ac731e901e967 (MD5) Previous issue date: 2017 | en |
| dc.description.tableofcontents | 口試委員審定書 ............................... i
誌謝 ....................................... ii 中文摘要 ................................... iii ABSTRACT ................................... iv CONTENTS .................................... v LIST OF FIGURES ........................... vii LIST OF TABLES ........................... viii Chapter 1 Introduction ...................... 1 1.1 Background and Motivation ............... 1 1.2 Workflow of Algorithm ................... 1 1.3 Contributions ........................... 2 1.4 Thesis Organization ..................... 2 Chapter 2 Related Work ...................... 3 2.1 Basic Method of Stereo Matching ......... 3 2.2 Some Cost Aggregation ................... 3 2.3 Using Bilateral Filter or Guided Filter . 4 2.4 Real-Time Stereo Matching ............... 5 Chapter 3 Implementation .................... 6 3.1 System Overview ......................... 6 3.2 Matching Cost ........................... 7 3.3 Cost Aggregation ........................ 9 3.4 Optimization ........................... 13 Chapter 4 Experiments and Results .......... 15 4.1 Implementation Environment ............. 15 4.2 Testing Data ........................... 16 4.3 Results ................................ 18 4.4 Discussion ............................. 19 4.5 Limitations ............................ 20 Chapter 5 Conclusion and Future Work ....... 23 5.1 Conclusion ............................. 23 5.2 Future Work ............................ 23 Bibliography ............................... 24 Resume ..................................... 26 | |
| dc.language.iso | zh-TW | |
| dc.subject | 引導影像濾波器 | zh_TW |
| dc.subject | 平行計算 | zh_TW |
| dc.subject | 立體匹配 | zh_TW |
| dc.subject | 開放計算語言 | zh_TW |
| dc.subject | 通用圖形處理器 | zh_TW |
| dc.subject | OpenCL | en |
| dc.subject | Stereo Matching | en |
| dc.subject | Parallel Computing | en |
| dc.subject | Guided Filter | en |
| dc.subject | GPU | en |
| dc.title | 以 OpenCL 實做使用引導影像濾波器的立體匹配演算法 | zh_TW |
| dc.title | OpenCL Implementation of Stereo Matching Algorithm with
Guided Filter | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 105-1 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 楊傳凱(Chuan-Kai Yang),葉正聖(Jeng-Sheng Yeh) | |
| dc.subject.keyword | 開放計算語言,立體匹配,平行計算,引導影像濾波器,通用圖形處理器, | zh_TW |
| dc.subject.keyword | OpenCL,Stereo Matching,Parallel Computing,Guided Filter,GPU, | en |
| dc.relation.page | 26 | |
| dc.identifier.doi | 10.6342/NTU201700034 | |
| dc.rights.note | 有償授權 | |
| dc.date.accepted | 2017-01-10 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 資訊網路與多媒體研究所 | zh_TW |
| 顯示於系所單位： | 資訊網路與多媒體研究所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-106-1.pdf 未授權公開取用 | 1.84 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。