應用於立體電視之DIBR虛擬視角映射技術與晶片設計

Chip Design and Implementation of Virtual View Generation System Based on DIBR Technology

沈德威(De-Wei Shen)

指導教授 ： 范育成

國立臺北科技大學/電資學院/電腦與通訊研究所/碩士(2013年)

摘要

隨著立體顯示器不斷推陳出新後，符合立體顯示器格式的內容物也大量被需求，其中包含裸眼式的多視角立體顯示器與眼鏡式的雙視角立體顯示裝置，其所需要內容物格式不盡相同，因此有相關研究提出DIBR (Depth-Image-Based Rendering) 虛擬視角映射技術，DIBR虛擬視角映射技術可以將原始視角搭配相關深度圖映射出多張虛擬視角以符合立體顯示器所需求的格式，但DIBR技術因系統本身不包含所有虛擬視角的影像資料，所以在合成虛擬視角時，物體邊緣很容易造成破洞(Hole)的現象，並造成使用者觀賞時的不適，因此為了將觀賞的舒適度提升，我們將致力改善影像品質。
本論文為了將影像品質提升，讓使用者有更好的立體視覺，因此會先將深度圖做前處理以減少深度圖中物件邊緣較不精確的部分，之後藉著視差調整讓較大破洞的部分重新切割成一個像素寬的細微破洞，最後我們將使用鄰近值填補的方法填補虛擬視角中的破洞部分，此外本論文將DIBR虛擬視角映射技術透過硬體架構設計來實現數位系統晶片，希望能藉由數位晶片來達成及時處理的效果並應用於消費性電子產品中。

關鍵字

多視角影像 ； 虛擬視角映射 ； 視差修改 ； 破洞填補

並列摘要

The contents which conform the stereoscopic display reach a great quantity demand when the stereoscopic display continuously innovates. Nevertheless, the formats of stereoscopic display are dissimilar. Therefore, some research proposed the DIBR (Depth-Image-Based Rendering) technology which could generate virtual views from reference view and associated depth map to match the format of stereoscopic display. However, the reference view doesn’t contain all information of virtual views. If we synthesized the virtual views, it would result in the hole region in virtual views. Consequently, it would make viewers uncomfortable when viewing stereoscopic images.
In order to comfort viewers’ perception, we endeavored to improve the quality of virtual view.
In the cause of obtaining the better 3D effect as well, we pre-processed the depth map to reduce the inaccurate edge of depth map. Then we introduced parallax modification algorithm to divide the enormous hole region into one-pixel-width holes. After parallax modification, we adopted the nearest interpolation algorithm and Telea inpainting algorithm to fill hole regions. Furthermore, we adopted hardware design to implement the DIBR system for real-time processing and applied in consuming electronic products as well.

並列關鍵字

Multi-view ； DIBR ； Parallax Modification ； Hole Filling

參考文獻

[8] T. M. Lehmann, C. Gonner, and K. Spitzer, “Survey: Interpolation Methods in Medical Image Processing,” IEEE Trans. on Medical Imaging, Vol. 18, No. 11, pp. 1049-1075, Nov. 1999.

[9] L. Zhang and W. Tam, “Stereoscopic Image Generation Based on Depth Images for 3DTV,” IEEE Trans. on Broadcasting, Vol.51, Issue 2, pp.191-199, June, 2005.

[11] L. Zhang, W. Tam, and D. Wang, “Stereoscopic Image Generation Based on Depth Images,” in IEEE International Conference on Image Processing (ICIP ‘04), vol. 5, pp. 2993-2996, Oct. 2004.

[12] W. Y. Chen, Y. L. Chang, S. F. Lin, L. F. Ding and L. G. Chen, “Efficient Depth Image Based Rendering with Edge Dependent Depth Filter and Interpolation,” in Proc. of IEEE International Conference on Multimedia and Expo, Amsterdam, July, 2005, pp. 1314-1317.

[13] P. J. Lee and Effendi, “Nongeometric Distortion Smoothing Approach for Depth Map Preprocessing”, IEEE Trans. on Multimedia, Vol. 13, pp. 246-254, Apr. 2011.