IEEE International Symposium on Circuits and Systems (ISCAS’2009) 24-27 May 2009, Taipei, Taiwan Efficient Inter Mode Decision for H.263 to H.264 Video Transcoding Using Support Vector Machines Speaker: Prof. Wan -Chi Siu Xuan Jing * , Wan-Chi Siu * , Lap-Pui Chau + and A.G. Constantinides ** *Department of Electronic and Information Engineering, Hong Kong Polytechnic University + School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore **Department of Electrical and Electronic Engineering, Imperial College London 1 Outline Introduction Inter Mode Decision in Transcoding Using SVM Experimental Results Conclusions 2 1
Introduction H.264 Interframe Mode Decision Inter block partition Candidate MB modes in P-frame SKIP, Intra-4 4, Intra-16 16, mode Inter-16 16, Inter-16 8, Inter-8 16, Inter-8 8 3 Introduction (cont.) Lagrangian RD cost function in H.264: J s c ( , ,mode| QP , ) SSD s c ( , ,mode| QP ) R s c ( , ,mode| QP ) mode mode The basic idea of fast mode decision in H.264 is to select the coding mode that achieves the best RD performance without searching all the modes , leading to complexity reduction. Objectives of proposed mode decision method for transcoding Making use of the information from H.263 decoding process to predict the H.264 Inter block coding mode for each MB. SVM-based mode prediction to early reject unlikely modes in order to achieve complexity saving 4 2
Inter Mode Decision in Transcoding Using SVM Proposed H.263 to H.264 transcoder with fast Inter mode decision 5 Inter Mode Decision in Transcoding Using SVM (cont.) Support vector machines (SVM) Based on a training set of examples with class labels and several attributes , the SVM constructs a decision rule which can be used predict the class lables of new input data given only attributes. In H.263 to H.264 transcoding: -- Class labels correspond to the optimal MB modes in H.264 -- Attributes correspond to extracted MB features from H.263 6 3
Inter Mode Decision in Transcoding Using SVM (cont.) Recall the SVM problem formulation: (e.g. binary classification) y y x {1, 1} ( , ) Given a training set of and a vector such that y i i i (to minimize the complexity (large margin) and training error ) 1 N w w T min 2 C i w ξ , , b C : Error penalty parameter i 1 N : total number of training data w is the normal vector perpendicular to w T x s.t. ( ( ) ) 1 , y b the hyperplane and i are slack i i i variables 0, 1,.... i N i Four extracted features from H.263 as attributes: Some terms MB mode in H.263 {Skip, Intra, Inter16x16, Inter8x8} Mean value of the MB residue Variance of the MB residue The number of non-zero AC coefficients of the MB, i.e. nzcoef 7 Inter Mode Decision in Transcoding Using SVM (cont.) SVM offline training process Based on free libSVM* software Four typical test sequences (Foreman, Stefan, Salesman, News) were firstly encoded and decoded by H.263 codec and then re-encoded by H.264 encoder Use RBF (radial basis function) kernel in SVM Some terms Use cross-validation to find the best penalty parameter, C, and the kernel parameter Cross-Validation * C.-C. Chang and C.-J. Lin, LIBSVM: a library for support vector machines. Software available at http://www.csie.ntu.edu.tw/~lin/libsvm 8 4
Inter Mode Decision in Transcoding Using SVM (cont.) SVM training results A few types of classification considered: -- Two-class classification {Skip, non-Skip} -- Three-class classification {Skip, Intra, Inter} -- Four-class classification {Skip, Intra, InterL, P8x8} (InterL includes Inter16x16, Inter16x8 and Inter8x16) Prediction accuracy of SVM classifier More than 80% accuracy is achieved. In our proposed transcoder, a few possible modes will be further added to the prediction mode list of the four-class SVM classifier output. 9 Inter Mode Decision in Transcoding Using SVM (cont.) Summary of the proposed Inter mode decision for transcoding 10 5
Experimental Results Comparison of Inter mode decision algorithms for transcoding The H.264 with full search mode decision is used as benchmark for comparison Average -0.02 0.37 -53.14 -0.03 0.61 -73.50 Alg in Literature 11 Experimental Results (cont.) R-D performance comparison Mother&Daughter Coastguard 12 6
Conclusions Proposed fast Inter mode decision for H.263 to H.264 video transcoding. Using selected features from H.263 to predict possible candidate modes in H.264 inter mode decision. SVM-based classifier is used for mode prediction Up to 77% complexity reduction is achieved with similar RD performance. Adv. & Disadv. 13 Thank You! 14 7
Recommend
More recommend