隨著立體影像顯示技術的進步，三維電視的應用也愈來愈普及。比起傳統只有左右眼兩種視角的三維影像，近年來發展出的多視角立體影像，勢必得耗費大量的頻寬和儲存設備，不利於三維電視的即時傳輸。基於深度圖像繪圖法(DIBR)便應運而生，將一張已知角度的影像和對應的深度圖，繪製出其他角度的圖片，再透過顯示器合成出立體影像。
使用此演算法，在繪製時可利用位移像素產生新的視角影像，但同時也伴隨著瑕疵，尤其在深度變化劇烈的地方會產生大的洞口，造成影像品質的降低。最常見的解決方法就是在前處理時將深度圖作二維平滑濾波，雖然有效，卻會讓立體效果衰減，造成幾何線條結構的扭曲，並增加電路設計的成本。
本論文將改善上述缺點，為了保有良好的立體效果，避免邊緣的瑕疵，故以遞減函數取代高斯濾波，針對深度圖中前景和背景的邊緣，把前景的像素作擴張(dilation)，改良彩色圖片及其深度圖在邊緣地方發生錯位的問題，再對背景的部分作遞減函數運算，縮減空洞的面積，降低空洞填補的複雜度。在電路設計中，消除位移算式中的除法，簡化成適用硬體的演算法。如此一來，空洞填補只須採用簡單的水平內插，即可達到良好的合成效果。最後提出一個多視角的DIBR立體影像合成器架構，在電路面積和處理時間達成平衡，使系統的運算速度能達到即時的應用。

The products of 3D stereoscopic vision are more and more popular owing to the improvement in 3D display technology. The multi-view three dimensional TV (3D-TV) is expected to be the next generation of stereoscopic television. It is impractical to transmit images of all viewpoints due to the limitations of bandwidth and storage. Depth image based rendering (DIBR) is a key technique to generate virtual views for different viewpoints from a known image and its corresponding depth map.
In this way, a major problem is disocclusion region (holes) exposed in the synthesized view. The big holes mainly appear in sharp transition of the depth map. Many algorithms have been presented to tackle the problems, and the most common one is to smooth the depth map before rendering. However, pre-filtering of the depth map usually bring other artifacts, such as geometric distortion and degradation of depth perception. Besides, the 2D filters would increase the cost of circuit design.
In order to maintain a good stereo effect and avoid geometric distortion, Gaussian filter is replaced by a descent function in this thesis. For the edge misalignments between a 2D image and its corresponding depth map, the foreground pixels are expanded to cover the transition region. And then the part of the background is modified by a descent operation to reduce the difficulty of hole-filling. In circuit design, we eliminate the division of the warping formula. Thus, the rendered views filled by a simple horizontal interpolation can obtain good synthesized effects. Finally, we propose a multi-view DIBR architecture for stereoscopic image generation and make a balance between circuit area and processing time. Experimental results show the proposed system can achieve real-time applications with slightly quality degradation.

[1] http://www.3d-display-info.com/tags/market-reports
[2] C. Fehn, “Depth-image-based-rendering (DIBR), compression and transmission for a new approach on 3D-TV,” in Proc. International Society for Optical Engineering Conf. Stereoscopic Displays and Virtual Reality Systems XI, San Jose, CA, vol.5291, pp.93–104, Jan. 2004.
[3] Y.-H. Chen, “Using Expanding Parallax Map to Preserve Edge Information for 3DTV,” Master’s Thesis, Dept. of Electrical Engineering, National Cheng Kung University, July 2012.
[4] C. Fehn, “A 3D-TV approach using depth-image-based rendering (DIBR),” in Proc. Vis., Imaging, and Image Process., Sept. 2003, pp. 482-487.
[5] L. Zhang and W.J. Tam, “Stereoscopic Image Generation Based on Depth Images for 3D TV,” IEEE Trans. On Broadcasting, vol. 51, No.2, pp. 1191-199, June 2005.
[6] J. Compen, “3D graphics rendering for multi-view displays: Using programmable shaders on graphics cards,” Technical note TN-2004/00920, Koninklijke Philips Electronics N.V. 2005.
[7] N. Holliman, “3D Display Systems,” Department of Computer Science, University of Durham, Science Laboratories, South Road, Durhamf, DH1 3LE, February, 2005.
[8] P. Benzie, J. Watson, P. Surman, I. Rakkolainen, K. Hopf, H. Urey, V. Sainov, and C. v. Kopylow, “A Survey of 3DTV Displays: Techniques and Technologies,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 17, No. 11, NOVEMBER 2007.
[9] N. A. Dodgson, “Autostereoscopic 3D Displays,” Computer, vol. 38, No. 8, pp. 31-36, August, 2005.
[10] Philips, Philips Multiview 3D Display Solutions,
http://www.infoterminals.eu/datasheet/philips_3d.pdf
[11] C. Fehn, P. Kauff, M. Op de Beeck, M. Ernst, W. IJsselsteijn, M. Pollefeys, L. Van Gool, E. Ofek, and I. Sexton, “ An Evolutionary and Optimized Approach on 3D-TV,” Proceedings of International Broadcast Conference, Amsterdam, The Netherlands, 2002.
[12] R. P. Berretty and F. Ernst, “High-Quality Images from 2.5D Video,” Proceedings of EUROGRAPHICS, Philips Research, Eindhoven, The Netherlands, September, 2003.
[13] X. Xu, L.-M. Po, K.-W. Cheung, K.-H. Ng, K.-M. Wong, and C.-W. Ting “A FOREGROUND BIASED DEPTH MAP REFINEMENT METHOD FOR DIBR VIEW SYNTHESIS,” Acoustics, Speech and Signal Processing (ICASSP), City Univ. of Hong Kong, 2012.
[14] W. J. Tam, G. Alain, L. Zhang, T. Martin, and R. Renaud, “Smoothing depth maps for improved stereoscopic image quality,” Proc. SPIE Conf. Three-Dimensional TV, Video, and Display III, Philadelphia, U.S.A., Vol. 5599, pp. 162-172, Oct. 2004.
[15] I. Daribo, C. Tillier and B. Pesquet-Popescu, “Distance Dependent Depth Filtering in 3D Warping for 3DTV,” Workshop on Multimedia Signal Processing, ENST, France, October, 2007.
[16] S.-B. Lee and Y.-S. Ho, “Discontinuity-adaptive depth map filtering for 3d view generation,” in IMMERSCOM ’09: Proceedings of the 2nd International Conference on Immersive Telecommunications, NO.8, ICST, 2009.
[17] C. Vázquez, W. J. Tam and F. Speranza, “Stereoscopic Imaging: Filling Disoccluded Areas in Depth Image-Based Rendering,” Proc. Of SPIE, Vol. 6392, Canada, 2006.
[18] K-J. Oh, S. Yea and Y-S. Ho, “Hole filling method using depth based in-painting for view synthesis in free viewpoint television and 3-d video,” in PCS, Piscataway, NJ, USA, 2009, pp. 233–236, IEEE Press.
[19] L.M. Po, S. Zhang, X. Xu, and Y. Zhu, “A new multidirectional extrapolation hole-filling method for depth-image-based rendering,” Processing of ICIP2011, pp. 2637-2640, Sep. 2011.
[20] C.-H. Chen, C.-Y. Sung, T.-Y. Chen, B.-T. Guo, and C.-H. Chen ”An Efficient Hole-Filling Approach Using Adaptive Rendering,” Genetic and Evolutionary Computing (ICGEC), pp. 336-339, Taiwan, August, 2012.
[21] H.-J. Chen, F.-H Lo, F.-C. Jan, and S.-D. Wu,“Real-Time Multiview Rendering Architecture for Autostereoscopic Displays,” Proc. IEEE Intl, Symp. Circuits and Systems (ISCAS), pp. 1165-1168, May 2010.
[22] H.-C. Chen “Real-time Stereoscopic Image Generation Using Depth Image Based Rendering with Virtual View Point Estimation,” Master’s Thesis, Dept. of Electrical Engineering, National Cheng Kung University, July 2010.
[23] W.-Y. Lee “Improved Hole-filling Algorithms and Hardware Implementation for Virtual View Generation,” Master’s Thesis, Dept. of Electrical Engineering, National Tsing hua University, July 2011.
[24] S.-F. Hsiao, J.-W. Cheng, W.-L. Wang, and G.-F. Yeh. Wu,“Low Latency Design of Depth-Image-Based Rendering Using Hybrid Warping and Hole-Filling,” Circuits and Systems (ISCAS), 2012 IEEE International Symposium on, pp. 608-611, May 2012.
[25] MSR 3D Video Download: http://research.microsoft.com/en-us/um/people/sbkang/3dvideodownload/