Video Bit stream Figure 1 a AVS China encoder 3 Video output Encoded bit stream Figure 1 b AVS China decoder 3 Inverse DCT Intra Mode decision Forward Frame Buffer Motion Compensation Figure 1 b 1 Correction to be made at the AVS China decoder showing the intra prediction block connected to Major blocks in AVS China 3 16 bit 8x8 integer transform Quantization and scaling and the scaling is preperformed in the encoder itself as compared to H 264 Intra prediction 5 modes Motion compensation 4 modes 8x8 8x16 16x8 16x16 Quarter pixel interpolation De blocking Entropy coding Minor blocks in AVS China 3 Motion vector prediction Adaptive scan Bit stream flow structure Sequence Frame Slice Macro Block Block Figure 1 c Bit stream flow structure 2 8X8 Integer cosine transform 2 AVS adopts a pre scaled Integer transform technique that is performed in the encoder itself The forward transform is given as 2 where f C stands for the input pixel matrix and the transformed coefficient matrix respectively After forward transform and combined scaling is performed 2 8x8 8x8 Figure 1 d 8x8 Integer cosine transform and the transformed coefficient matrix 2 Different prediction and compensation modes Intra frame prediction mode 2 Spatial prediction technique is implemented in AVS China The intra prediction is based on 8x8 block structure AVS China implements 5 luminous intra prediction technique and 4 chrominance intra prediction technique The reconstructed pixels of neighboring block before the De blocking filtered is used as a reference pixels to current block Figure 1 e Neighboring pixels in luminance intra prediction 3 Figure 1 f Five different luminance intra prediction modes 3 Inter frame prediction 3 There are 4 macro blocks specification for the inter frame prediction namely 8x8 8x16 16x8 16x16 In p picture there are 5 inter prediction modes namely P skip 16x16 P 8x8 P 8x16 P 16x8 P 16x16 For the last 4 modes of inter prediction each partition of macro block is predicted from one of the two reference frames which are latest decoded I or P frame There are two kinds of Bi Predictions in AVS China namely the symmetric and the direct prediction In symmetric only one forward motion vector is transmitted for each partition In direct forward and backward motion vectors are derived from motion vector of the collocated block Figure 2 a Symmetric mode in AVS China 1 Interpolation 1 A quarter pixel interpolation named as Two Step four Tap TFST method is implemented in AVS China pixel samples are interpolated in step 1 and pixel samples are interpolated in step 2 pixel Interpolation method is a 4 Tap filter H1 1 8 5 8 5 8 1 8 For ordinary pixel samples a c d f I k n and q in the below shown Figure 2 b 4 Tap filter H2 1 16 7 16 7 16 1 16 is applied and four special pixel samples e g p and r are filtered by two Tap bilinear filter H3 1 2 1 2 Figure 2 b Position of integer pixels pixels and pixels wherein a c d f l k n and q are ordinary pixels e g p and r are special pixels 1 Performance analysis of AVS China 6 In MPEG 1 MPEG 2 MPEG 4 the transform block size is 8x8 and their smallest motion compensation block size are 16x16 16x8 and 8x8 respectively In H 264 AVC the smallest transform and intra and inter compensation block size are of 4x4 The smaller the block size the more precise is the inter and intra compensation and the smaller the residuals are The smaller the block size is more bits are required for motion vectors and intra prediction modes Hence a series of experiment which compares the coding efficiency of block sizes 4x4 and 8x8 have been tabulated Figure 2 c Experimental results showing the performance comparison between AVS China and H 264 AVC 6 Figure 2 d showing the PSNR gain at the same bit rate of 8x8 block size in AVS China over the 4x4 block size in H 264 and the bit rate increase at the same PSNR of the 8x8 block size in AVS China over the 4x4 block size in H 264 for different sequences 6 Figure 3 a comparison of AVS China with MPEG 2 and H 264 AVC 4 6 Figure 3 b comparison of AVS China H 264 AVC and MPEG 2 on HD sequence 4 6 An overview on VP6 coding technique 7 VP6 is the coding technique developed by On2 Technologies Flash Media is emerging as the new preferred solution over the existing Windows Media Player Apple QuickTime and Real Network Player for providing video services over the Internet Macromedia adopted VP6 as their video coding algorithm for its Flash player in 2005 7 VP6 on Flash8 provides better performance than the existing standards with smoother streaming and low color contrast video images 7 Hence forth this creates an opening for developing an AVS to VP6 transcoder to transmit the AVS coded signals over the internet Proposed Research Based on the available reference materials like H 263 to VP6 transcoder 8 AVS to MPEG 2 transcoder 5 performance comparison of AVS and H 264 AVC video coding standard 6 a new method can be proposed to develop the AVS to VP6 transcoder as shown in the figure 4 a Video AVS China encoder AVS China to VP6 transcoder Internet Flash Media VP6 VP6 decoder Figure 4 a showing the implementation of AVS to VP6 transcoder References 1 L Yu et al An Overview of AVS Video tools performance and complexity Visual Communications and Image Processing 2005 Proc of SPIE vol 5960 pp 596021 July 31 2006 2 Zhang L Yu An area efficient VLSI architecture for AVS intra frame encoder Visual Communications and Image Processing 2007 Proc of SPIE IS T Electronic Imaging SPIE vol 6508 pp 650822 Jan 29 2007 3 W Gao et al AVS The Chinese Next Generation Video Coding Standard NAB Las Vegas 2004 4 T Wiegand J Sullivan Overview of the H 264 AVC Coding Standard IEEE Trans Circuits Syst Video Technol vol 13 pp 560 576 July 2003 5 J Wang et al An AVS to MPEG2 Transcoding System China Proceedings of 2004 International Symposium on Intelligent Multimedia Video and Speech Processing Hong Kong pp 302 305 October 20 22 2004 6 X Wang D Zhao Performance comparison of AVS and H 264 AVC video coding standards J Comput Sci Technol Vol 21 No 3 pp 310 314 J May 2006 7 On2 Technologies Inc WHITE PAPER On2 VP6 for Flash8 Video Sept 12 2005 8 C Holder and H Kalva H 263 to VP6 transcoder SPIE vol 6822 VCIP pp 68222B San Jose CA Jan 2008 Web References AVS China software 1 http www avs org cn en 2 http www eetimes com news latest showArticle jhtml articleID 199900715 3 ftp 159 226 42 57 public avs doc avs software THANK YOU
View Full Document
Unlocking...