以移動向量特性於多視角視訊之錯誤補償決策演算法

Using Motion Correlation in Error Concealment Method Selection for Multi-view Video Coding

嵇正言(Cheng -Yen Chi)

指導教授 ： 李佩君

國立暨南國際大學/科技學院/電機工程學系/碩士(2011年)

摘要

近年來多媒體服務蓬勃發展，為提供更真實的場景如同人眼所見的效果，發展出許多3D視訊技術，在這些技術當中又以多視角視訊(Multi-view video，MVV)技術最令人關注，主因為多視角視訊提供高品質的3D視覺效果，其相關應用如3DTV、自由視角電視系統(Free-Viewpoint Television ，FVT)，觀賞者可在同一場景中選擇不同的視角來觀看，使3D觀賞感受更加的真實。這些不同的視訊序列(sequence)是由多個攝影機，相同時間與場景以不同拍攝角度拍攝所獲得。然而，多個視訊序列雖提昇了3D觀賞效果卻需更大的資料量，造成傳輸的負擔更加沉重，於是多視角視訊壓縮系統(Multi-view View Coding System)中視訊壓縮的技術就必須在不影響觀賞者觀看的影像品質下，盡可能提高壓縮率以減少傳輸的負擔和降低儲存資料量，多視角視訊編碼其中利用空間性(視角間)、時間性(相同視角)的畫面關係的相依性提昇多視角序列的壓縮率，因此每一個被壓縮的bit內的資訊都可能與視角間和相同視角相關，如果在傳輸過程中發生錯誤將會遺失掉更多的資訊，所造成的錯誤延續效應不單只影響相鄰的畫面甚至蔓延至相鄰的視角，再由相鄰的視角影響相鄰視角的相鄰畫面，其影響程度比單一視角更為劇烈，嚴重破壞解碼端視訊品質，因此就必須在解碼端進行錯誤修補的動作，降低傳輸錯誤的影響，本論文針對多視角中主視角與輔助視角提出一個判斷方式可以由鄰近的資訊判別出遺失資訊中區塊的模式，根據不同的區塊模式給予不同的移動向量候選區域並利用邊界匹配在在多視角支援的參考方向中，重建移動向量與參考方向，但是錯誤修補選擇不同區塊模式及參考方向，雖然可以提升畫面修復品質，同時也增加了計算複雜度，因此本論文再提出利用周圍巨區塊移動量特性選擇不同修補方式的錯誤修補方法減少視訊品質的下降同時有效降低計算複雜度。模擬結果顯示，本論文提出的方法整體PSNR約可提高0.5~6.4dB。

關鍵字

H.264/AVC ； 多視角視訊編碼 ； 視訊傳輸 ； 錯誤修補

並列摘要

With the rapid development of multimedia technology, the demand for the realistic 3D effect of observed scenes have become available. Multi-video is a key technology for stereo video that provides the user with a new viewing experience. It includes many applications such as 3DTV, free viewpoint television (FVT). As watch programs users can choose different view at the same scene. The multi-view sequences utilize several cameras to capture a scene at the same time by different angle. The data of 3D image are much huger than that of 2D image and need lager transmission bandwidth. Therefore, an efficient compression algorithm is needed to compress the huge data and reduce the required bandwidth. The coding scheme exploits spatial, temporal and inter-view correlation to achieve the high compress efficiency. Multi-view Video Coding (MVC) easy causes the error propagation when transmission error occurs. Hence, error concealment method is employed to enhance the video quality.
This paper proposed variable block mode prediction Interview/temporal EC　algorithms for improving the video quality. First, it uses the correlation of block mode between the lost macroblock (MB) and its neighboring MB to predict the possibly mode of the lost MB. Then, the lost block mode has been obtained. Second, this paper considers the variable mode to decide candidate Motion Vectors/Disparity Vectors that exploits the corresponding Motion information in neighboring MB. The proposed scheme uses the boundary matching to evaluate the Sum of Absolute Difference (SAD) from candidates. Finally, it compares the SAD each reference directions which owns minimum value. The lost reference directions and MV/DV can be reconstructed. But the boundary matching spends computational complexity to help concealment accurately. Therefore, this paper proposed the third algorithm using the motion correlation in surrounding MB in selection method to conceal the corrupted video. Experimental result shows the proposed method not only improves the quality of video but also reduces the computational complexity. The proposed method have PSNR with 0.5~6.4dB improvement more than the other traditional method.

並列關鍵字

H.264/AVC ； multi-video coding ； video transmission ； error concealment

參考文獻

[1] H. M. Ozaktas and L. Onural, Three-dimensional Television, Capture, Transmission, Display, USA: Springer, 2008.

[2] P. Merkle, A. Smolic, K. Muller, and T. Wiegand, “Efficient prediction structures for Multiview Video Coding”, IEEE Transactions on Circuits and Systems for Video Technology, vol. 17, no. 11, pp. 1461-1473, Nov. 2007.

[3] A. Smolic, K. Mueller, N. Stefanoski, J. Ostermann, A. Gotchev, G. B. Akar, G. Triantafyllidis and A. Koz, “Coding Algorithms for 3DTV—A Survey”, IEEE Transactions on Circuits and Systems for Video Technology, vol. 17, no. 11, pp. 1606-1621, Nov. 2007.

[4] P. Benzie, J. Watson, P. Surman, I. Rakkolainen, K. Hopf, H. Urey, V. Sainov and C. von Kopylow, “A Survey of 3DTV Displays: Techniques and Technologies”, IEEE Transactions on Circuits and Systems for Video Technology, vol. 17, no. 11, pp. 1647-1658, Nov. 2007.

[5] G. B. Akar, A. M. Tekalp, C. Fehn, and M. R. Civanlar, “Transport Methods in 3DTV—A Survey”, IEEE Transactions on Circuits and Systems for Video Technology, vol. 17, no. 11, pp. 1622-1630, Nov. 2007.