Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/6977
標題: 應用於3DAV標準之高畫質視合成演算法設計及其電路架構設計與實現
High Quality View Synthesis Algorithm and Architecture Design for 3DAV Standard
作者: 林君瑋
Lin, Jung-Wei
關鍵字: 3DAV Standard;3DAV標準;Synthesis Algorithm;視訊合成
出版社: 電機工程學系所
引用: [1] Christoph Fehn, "A 3D-TV approach using depth-image-based rendering (DIBR)", in Proc. Visual. , Image., Image Process., pp. 482-487, sep. 2003. [2] Liang Zhang and Wa James Tam, “Stereoscopic image generation based on depth images for 3D TV”, IEEE Trans. Broadcast., vol. 51, no. 2, pp.191-199, Jun. 2005. [3] Karsten Müller, Aljoscha Smolic, Kristina Dix, Philipp Merkle, Peter Kauff, and Thomas Wiegand, “View synthesis for advanced 3D video systems”, EURASIP J. Image Video Process., vol. 2008, pp. 1-11, 2008, Article ID 438148. [4] Dong Tian, Po-Lin Lai, Patrick Lopez, and Cristina Gomila, “View synthesis techniques for 3D video”, in Proc. SPIE 7443, 2009. [5] Yuji Mori, Norishige Fukushima, Tomohiro Yendo, Toshiaki Fujii, and Masayuki Tanimoto, “View generation with 3D warping using depth information for FTV”, Signal Process.: Image Commun., vol. 24, no. 1-2, pp. 65-72, Jan. 2009. [6] M. Domanski, M. Gotfryd, and K. Wegner, “View synthesis for multiview video transmission”, in Proc. Int. Conf. Image Process., Comput. Vision Pattern Recog., 2009, pp. 13-16. [7] Marcelo Bertalmio and Guillermo Sapiro, “ Image inpainting”, SIGGRAPH 2000, pages 417 - 424. [8] Andre Redert, Marc Op de Beeck, Christoph Fehn, Wijnand IJsselsteijn, Marc Pollefeys, Luc Van Gool, Eyal Ofek, Ian Sexton, Philip Surman, “ATTEST: advanced three-dimensional television system techniques”, Proc. of 3DPVT'02, pp. 313-319, Jun. 2002. [9] D. McCutchen A. Smolic, “Applications and requirements for 3dav”, in ISO/IEC JTC1/SC29/WG11, N5877, July 2003. [10] Wan-Yu Chen, Yu-Lin Chang, Shyh-Feng Lin, Li-Fu Ding, and Liang-Gee Chen, “Efficient depth image based rendering with edge dependent depth filter and interpolation”, IEEE International Conference on Multimedia and Expo, pp. 1314 - 1317, July 2005. [11] Wan-Yu Chen, Yu-Lin Chang, and Liang-Gee Chen, “Realtime depth image based rendering hardware accelerator for advanced three dimensional television system”, Proceedings of IEEE International Conference on Multimedia and Expo, pp. 2069 - 2072, July 2006. [12] Liang Zhang and Wa James Tam, “Stereoscopic image generation based on depth images for 3D TV”, IEEE Transactions on Broadcasting, vol. 51, no. 2, pp. 191 - 199, June 2005. [13] Karsten Muller, Philipp Merkle, and Thomas Wiegand, “3-D video representation using depth maps”, Proceedings of the IEEE, vol. 99, no. 4, pp. 643 - 656, April 2011. [14] Pei-Yin Chen, Chien-Chuan Huang, Yeu-Horng Shiau, and Yao-Tung Chen, “A VLSI implementation of barrel distortion correction for wide-angle camera images”, IEEE Trans. Circuits Syst. II: Express Briefs, vol. 56, no. 1, pp. 51 - 55, Jan 2009. [15] Hau T. Ngo, and Vijayan K. Asari, “A pipelined architecture for real-time correction of barrel distortion in wide-angle camera images”, IEEE Trans. Circuits Syst. Video Technol., vol. 15, no. 3, pp. 436-444, Mar. 2005. [16] Pin-Chih Lin, Pei-Kuei Tsung, Liang-Gee Chen, “Low-cost hardware architecture design for 3D warping engine in multiview video applications”, Proceedings of IEEE International Symposium on Circuits and Systems 2010, pp. 2964-2967, May 2010. [17] Pei-Kuei Tsung, Pin-Chih Lin, Li-Fu Ding, Shao-Yi Chien, and Liang-Gee Chen, “Single iteration view interpolation for multiview video applications “, Proceedings of the IEEE 3DTV Conference, pp. 1-4, May 2009. [18] Seok-Hoon Kim, Jae-Sung Yoon, Chang-Hyo Yu, Donghyun Kim, Kyusik Chung, Han Shin Lim, Yun-Gu Lee, HyunWook Park, Jong Beom Ra, and Lee-Sup Kim, “A 36 fps sxga 3-d display processor embedding a programmable 3-d graphics rendering engine”, IEEE Journal of Solid-State Circuits, vol. 43, no. 5, pp. 1247 - 1259, May 2008. [19] Seok-Hoon Kim, Hong-Yun Kim, Young-Jun Kim, Kyusik Chung, Donghyun Kim, and Lee-Sup Kim, “A 116fps 74mw mobile heterogeneous 3d-media processor for 3d display contents”, Symposium on VLSI Circuits, pp. 258 -259, June 2009. [20] Furo-cho, Chikusa-ku, Nagoya, ”Overview of FTV (Free-Viewpoint Television)”, IEEE International Conference on Multimedia and Expo, pp.1552-1553, June 2009. [21] Eero Aho, Jarno Vanne, Kimmo Kuusilinna, and Timo D. Hamalainen, “Comments on “Winscale: an image-scaling algorithm using an area pixel model””, IEEE Transactions on Circuits and Systems for Video Technology, vol. 15, no. 3, pp. 454 - 455, March 2005. [22] Pei-Kuei Tsung, Ping-Chih Lin, Kuan-Yu Chen, Tzu-Der Chuang, Hsin-Jung Yang, Shao-Yi Chien, Li-Fu Ding, Wei-Yin Chen, Chih-Chi Cheng, Tung-Chien Chen, Liang-Gee Chen, “A 216fps 4096×2160p 3DTV set-top box soc for free-viewpoint 3DTV applications”, in IEEE International Solid-State Circuits Conference, pp. 124-126, April 2011.
摘要: 
影像視頻在休閒娛樂上一直佔有著相當重要的角色,為追求更高品質的視覺享受,影像的顯示不斷地往更高解析度的方向發展,如HD 720p、Full HD 1080p,到將來更高階的QFHD顯示規格,但僅以2D的顯示方式較難以讓觀賞者有身歷其境的感受,因此3D視頻便成為一熱門的研究主題。
為使觀賞者有3D立體的實境感,在顯示上是讓左、右眼分別看到其相對應之畫面,再由大腦合成出立體的畫面。而左、右眼影像的獲得,可直接在拍攝時便錄製,如此在顯示上會有較高的畫面品質;若拍攝時僅採用一般的2D攝影方式,便需要經過2D轉3D的過程,來產生左、右眼的畫面,在MPEG-2中定義了DIBR這種2D轉3D影像的作法(3DAV-Advanced 3D TV),其中視點合成的部份即為本論文所探討的主題。
3D的視點合成演算法分成三個步驟,針對深度圖的預處理、影像Warping,以及空洞填補,其主要目的為解決經由Warping後所產生之空洞,藉由深度圖的預處理可儘量減少Warping後畫面的空洞,以減輕空洞填補時的困難。
所提出的3D視點合成演算法針對各流程優缺點作分析,改良原先Warping運算,加入影像差補點之概念,提昇Warping畫面品質,並針對不同位置之空洞產生作分析,選擇最適切的填補作法,得到最後的合成畫面。
硬體部份針對高解析度Full HD 1080p及60fps之高顯示頻率影像規格作設計,採用TSMC 0.18μm製程,並對演算法各步驟之精確度問題、硬體成本作化簡及數據分析,以期達到高畫質且實時顯示的目的。

Video always plays an important role in entertainment. There is a need to pursue high quality videos. For that reason the display of images continues to develop higher resolution, such as HD 720p, Full HD 1080p, and QFHD for future display spec. However, the 2D video display method is difficult for viewers to have virtual reality. 3D video, therefore, becomes a popular topic in recent years.
The virtual reality with 3D stereo videos is created by having the right eye and left eye seeing the right image and left image separately. Then human brain will synthesize the stereo feeling for us. To obtain the right and left image, one method is to capture it directly. This method will provide us higher quality in display. If we capture the video in 2D video's method, we then need to process the video by some steps in order to generate the right and left images. In the definition of MPEG-2, it describes the view synthesis method of DIBR(3DAV-Advanced 3D TV), and it is the research topic of this paper.
The algorithm of 3D view synthesis can be separated by three steps: preprocessing of depth map, image warping, and hole filling. The main purpose of view synthesis is to fill the holes generated by image warping. By preprocessing the depth map, we could reduce the size of holes, and relieves the difficulty of hole filling.
The proposed 3D view synthesis algorithm analyzes the three steps of original view synthesis flow and adds the concept of image interpolation to improve the quality of original warping computations. We analyze the holes in different locations in one frame at the same time, and choose the most appropriate method to fill these holes.
The implementation of the chip by TSMC 0.18μm technology file can generate the spec of Full HD 1080p, and 60fps of its frame rate. We also analyze the fractional problem and minimize the hardware requirements. By doing so, we wish to obtain real time display with high quality.
URI: http://hdl.handle.net/11455/6977
其他識別: U0005-2112201114375500
Appears in Collections:電機工程學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.