Implementation and Performance Analysis of 2-D Order 16 Integer Transforms in H.264/AVC and AVS-video for HD video codingOutlineDiscrete Cosine Transforms (ICT)Pros and ConsDevelopment of ICTH.264/AVC encoderH.264/AVC decoderICT in H.264/AVCSlide 9Slide 10Slide 11Slide 12AVS-video encoderAVS-video decoderICT in AVS-videoSlide 16Slide 17Order 16 ICTOrder 16 ICT and HD video codingSlide 20Simple order 16 ICT (SICT)Slide 22Simple order 16 ICTSlide 24Modified order 16 ICTSlide 26Slide 27Slide 28Slide 29Slide 30Order 16 binDCT-LSlide 32Slide 33Implementation in H.264/AVCImplementation in AVS-videoTransform Coding gainSlide 37Slide 38SICT in H.264/AVC (1280 x 720)MICT in H.264/AVC (1280 x 720)binDCT-L in H.264/AVC (1280 x 720)SICT in AVS-video (1280 x 720)MICT in AVS-video (1280 x 720)binDCT-L in AVS-video (1280 x 720)Vidyo1(1280 x 720)ConclusionsReferencesSlide 48Slide 49Slide 50Implementation and Performance Analysis of 2-D Order 16 Integer Transforms in H.264/AVC and AVS-video for HD video coding Madhu Peringassery KrishnanMultimedia Processing Lab, University of Texas at Arlington, TX, USA.Advisor: Dr. K. R. RaoOutline•Discrete Cosine Transform (DCT-II)•Development of Integer Cosine Transform (ICT)•ICT in H.264/AVC•ICT in AVS-video •Order 16 ICT•2-D order 16 ICT and HD video coding•Simple order 16 ICT (SICT)•Modified order 16 ICT (SICT)•binDCT-L•Implementation•Performance Analysis•Conclusions•ReferencesDiscrete Cosine Transforms (ICT) • Discrete Cosine Transform (DCT-II) [1]-- k, n = 0,1,2,……..,N-1• 2-D transform is separable into two 1-D transforms evaluated along rows followed by columns [2].Pros and ConsDCT-II •Pro : Good energy compaction capability Fast algorithms for implementation•Con : Involves floating-point arithmetic Mismatch between forward and inverse transformICT [3]•Pro : Integer arithmetic implementation Avoid mismatch between forward and inverse transform Good energy compaction capability if well designed Fast algorithms can be developed •Con : Orthogonality depends on the elements of transform matrixDevelopment of ICT•Approximation of DCT-II- [3]where k is scaling factor and T is ICT•Elements of T [3]- Maintain relative magnitude and signs- Posses dyadic symmetry [4]- OrthogonalityH.264/AVC encoderTypical block diagram of a H.264/AVC encoder [5]H.264/AVC decoderTypical block diagram of a H.264/AVC decoder [5]ICT in H.264/AVCICT•Order 4 ICT [6]•Order 8 ICT [7]Other transforms: •4 × 4 Hadamard transform applied to the DC coefficients of 4 × 4 integer transforms (intra-predicted 16 x 16 macroblocks) •Additional 2 × 2 Hadamard transform applied to DC coefficients of 4 × 4 integer transforms for chroma componentsICT in H.264/AVC•4 x 4 ICT matrix- •4 x 4/2 x 2 Hadamard matrix-ICT in H.264/AVC•8 x 8 ICT matrix- •Non-normalized•Fast implementation [8]ICT in H.264/AVC•Flow diagram for 8 x 8 ICT [9]-ICT in H.264/AVC•Sparse matrix factors : whereAVS-video encoderTypical block diagram of AVS-video encoder [10]AVS-video decoderTypical block diagram of AVS-video decoder [10]ICT in AVS-video•Order 8 ICT [11]-•Order 16 ICT : extended from order 8 ICT•Fast implementationICT in AVS-video•Flow diagram for 8 x 8 ICT [9]-ICT in AVS-video•Sparse matrix factors : whereOrder 16 ICT•Approximated from order 16 DCT-II- [12]•General transform matrix [13] - ‘E’ denotes even symmetry and ‘O’ denotes odd symmetry about the solid line (mirror image and negative of mirror image)Order 16 ICT and HD video coding•Spatial correlation of HD videos are higher- [14] where E is the ensemble average operator x(n1) and x(n2): intensity values of n1,n2 1,2: mean 1,2: standard deviation•Better coding efficiency using higher order transformsOrder 16 ICT and HD video codingTable showing spatial correlation of prediction errorTest sequences Resolutionr(1) r(2)Mean Standard DeviationMean Standard DeviationKimono1920 × 1080 (HD)0.8673 0.1284 0.7311 0.1434Parkscene 0.7431 0.1820 0.6695 0.1967Cactus 0.8542 0.1692 0.7483 0.1245Vidyo11280 × 720 (HD)0.7539 0.2401 0.4073 0.1842Vidyo2 0.6643 0.1982 0.3060 0.1569Vidyo3 0.5474 0.1125 0.3221 0.2923PartyScene832 × 480 (WVGA)0.4953 0.1598 0.2019 0.1757BQMall 0.4517 0.2145 0.1966 0.2450BasketballDrill 0.5594 0.1183 0.2301 0.1032BQSquare416 × 240 (WQVGA)0.3543 0.2935 0.0964 0.1722BlowingBubbles 0.2879 0.1515 0.0473 0.1906BaketballPass 0.2177 0.1784 0.0355 0.2098• Prediction error : Difference between original and intra or inter predicted macroblocksSimple order 16 ICT (SICT)•Extension of order 8 ICT [15]•Low complexity•Comparable transform coding gain with DCT-II (Plot 1)•Transform matrix of order 16 SICT for AVS-video • Requires 24 shifts and 88 additionsSimple order 16 ICT (SICT)•Transform matrix of order 16 SICT for H.264/AVC • Requires 20 shifts and 80 additionsSimple order 16 ICT•Flow diagram 16 x 16 SICT [15]Simple order 16 ICT•Sparse matrix factors :H.264/AVC : where AVS-video : wherewhere and order of input as shown in flow diagramModified order 16 ICT•Low complexity (more complex than SICT)•Comparable transform coding gain (better than SICT)•Steps involved in development [9]- Order 8 ICT of H.264/AVC or AVS-video is borrowed as the even part (T8e)- Modified dyadic symmetry of odd part of order 16 DCT-II symmetry (M8o) [9]Modified order 16 ICT•Transform matrix of order
View Full Document