H.264/AVC是近年來新一代的視訊壓縮標準，它具有更高的壓縮效能及更好的壓縮品質，H.264/AVC在做移動估測(Motion Estimation)時採用了包括16×16、16×8、8×16、8×8、8×4、4×8、4×4，及一個略過(Skip Mode)之可變區塊大小模式(Variable Block Size Mode)的選擇。並最多可參考到往前及往後五張畫面，以提供更加精準之變動預測。在一般影片中的連續兩個畫面中，利用移動補償(Motion Compensation)作移動估測之區域，在可變區塊大小模式上是具有相關性的。因此新近在H.264/AVC錯誤隱藏(Error Concealment)之研究有些是利用可變區塊模式對錯誤區塊作判斷，然後利用不同模式的候選區塊作時間域(Temporal)之錯誤隱藏。
基於H.264/AVC可變區塊大小模式下，本論文為了增進效率，提出了一個可變區塊模式預測之空間/時間域錯誤隱藏演算法(Variable Block Mode Prediction Spatial/Temporal Algorithm)，利用遺失巨區塊(Lost Macroblock)與鄰近之巨區塊的模式具相關性之特點，用以預測遺失巨區塊之性質。若預測之遺失巨區塊被判斷為略過(Skip)、16×16、 16×8、8×16或8×8等不同之模式，則針對各種不同之模式運用不同的時間域錯誤搜尋法加以修補。尤其是對於16×16之模式性質選用空間域或時間域錯誤隱藏法來修補。本論文所提出之利用H.264/AVC鄰近之移動向量做預測之特點以判斷何時使用空間域或時間域之錯誤隱藏法，對於計算速度及確保畫面的品質將有較佳的效果。
實驗結果顯示，本論文所提出的可變區塊模式預測之空間/時間域錯誤隱藏法，與適應性區塊大小演算法相比較證明其可提升僅在可變區塊大小模式作時間域錯誤隱藏之效率，而在影像品質上亦能保持甚佳之品質。

H.264/AVC is the newest video coding standard recently, and it performs more better in compression efficiency and compression quality. H.264/AVC supports seven different variable block size modes for a macroblock: 16×16、16×8、8×16、8×8、8×4、4×8、4×4 and a skip mode when using motion estimation. H.264/AVC uses up to five reference frames forward and backward to help the prediction accurately. Between the two successive frames in a video, the region of motion estimation after motion compensation is correlative with the variable block size mode. Some of the recently researches for error concealment on H.264/AVC use variable block mode to make a decision for error blocks, then they do temporal error concealment by using candidate blocks with different modes.
Based on the variable block size mode, we propose a variable block mode prediction spatial/temporal algorithm for improving the efficiency in this paper. The proposed scheme uses the correlation of modes between the lost macroblock and its surrounding macroblocks to predict the feature of the lost macroblocks. If the mode is skip、16×8、8×16 or 8×8, then we adopt different kinds of temporal error concealment according to its corresponding mode to recover the lost blocks. If the mode is 16×16, then we adopt spatial or temporal error concealment to recover the lost blocks. In this paper, the proposed scheme using the motion vectors of neighbor macroblocks to predict in H.264/AVC to decide when to use spatial or temporal error concealment will have a better performance on executing time and the image quality.
Experimental results are presented to demonstrate that the proposed scheme in this paper proves the efficiency in only using temporal error concealment based on variable block mode, and maintains the better image quality compared to adaptive block size algorithm.

[1] Bin Chen, Rong Luo, “An Adaptive Error Concealment
Scheme for H.264/AVC Transmission over Packet-Loss
networks,” Advanced Communication Technology, The 9th
International Conference on Volume 3, 12-14 Feb. 2007
Page(s):1880 - 1885

[2] C. Liu, H. Chen and Y. Zhao, “Temporal Error
Concealment Based on Variable MB Mode in H.264 Video
Transmission over Error-Prone Channels,” IEEE
Conference on Intelligent Control and Automation, 2006,
Volume 2, 21-23 June 2006 Page(s):9655- 9659.

[3] D. KIM, J. KIM, and J. JEONG, “Temporal Error
Concealment for H.264 Video Based on Adaptive Block-
Size Pixel Replacement,”IEICE Transactions on
Communications 2006, Volume E89-B, Number 7, July 2006,
Page(s):2111-2114.

[4] F. Chiaraluce, L. Ciccarelli, E. Gambi, S. Spinsante,
“Performance Evaluation of Error Concealment Techniques
in H.264 Video Coding,” International Symposium on
Picture Coding 2004, San Francisco, CA, Dec. 2004,
Paper 59.

[5] http://iphome.hhi.de/suehring/tml/

[6] Iain E. G. Richardson, “H.264 and MPEG-4 Video
Compression, “ WILEY.

[7] M. GHANBARI, “The Cross-Search Algorithm for Motion
Estimation, “ IEEE Transactions on Communications,
Volume 38, Issue 7, July 1990, Page(s): 950-953.

[8] Panos Nasiopoulos, Lino Coria-Mendoza, Hassan Mansour,
Adarsh Golikeri, “An Improved Error Concealment
Algorithm for Intra-frames in H.264/AVC,” IEEE
International Symposium on Circuit and Systems ISCAS
2005, Kobe, Japan. May 2005 Page(s):320-323.
[9] S. Aign, and K. Fazel, “Temporal & Spatial Error
Concealment Techniques for Hierarchical MPEG-2 Video
Codec,” IEEE International Conference on
Communications and Gateway to Globalization, 1995.
Volume 3, June 1995 Page(s):1778-1783.

[10] T Wiegand(editor), “Joint Final Committee Draft
(JFCD) of Joint Video Specification (ITU-T Rec. H.264
| ISO/IEC 14496-10 AVC), “Joint Video Team (JVT) of
ISO/IEC MPEG and ITU-T VCEG July 2002.

[11] T. Wiegand, G.J Sullivan, G. Bjontegaard, A. Luthra,
“Overview of the H.264/AVC Video Coding Standard,”
IEEE Transactions on Circuits and Systems for Video
Technology, , Volume 13, July 2003, Page(s):560-576.

[12] W. M. Lam, A. R. Reibman, and B. Liu, “Recovery of
lost or erroneously received motion vectors,” IEEE
Conference on Acoustics, Speech, and Signal Processing
1993, Minnesota, USA, Volume 5, April 1993, Pag(s):417-
420.

[13] W.Zeng and B.Liu: “Geometric-Structure-Based Error
Concealment with Novel Applications in Block-Based Low-
Bit-Rate Coding, “ IEEE Transactions on Circuits and
Systems for Video Technology. Volume 9, Issue 4, June
1999, Page(s): 648-665.

[14] Y.K. Wang, M.M. Hannuksela, V. Varsa, A. Hourunranta,
and M. Gabouj, “The Error Concealment Feature in the
H.26L Test Model,” International Conference on Image
Processing 2002, Volume 2, Sept 2002, Page(s):729-732.