OBJECT-BASED ANALYSIS-SYNTHESIS CODING OF IMAGE SEQUENCES AT VERY-LOW BIT RATES

被引:66
作者
GERKEN, P
机构
[1] Institut fur Theoretische Nachrichtentechnik und Informationsverarbeitung, Universitat Hannover, D-30167, Hannover
关键词
D O I
10.1109/76.305868
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
An object-based analysis-synthesis image sequence coder for transmission bit rates between 8 and 16 kbit/s is presented. Each moving object is described by three sets of parameters defining its shape, motion, and colour. Coding is based on the source model of flexible 2-dimensional objects which move translationally in the image plane, as it has been used in an implementation for 61 kbit/s ISDN videophone by Hatter. In order to cut down the bit rate from 64 kbit/s to 8 kbit/s, QCIF image resolution instead of CIF resolution is applied. Image analysis and coding of object parameters have been adapted to the reduced resolution and to the changed parameter statistics, respectively. In addition to Hotter's coder, predictive coding is used for encoding polygons and splines to improve the coding efficiency of shapes. Vector quantization is applied instead of DCT for coding the luminance and chrominance parameters of the object textures. Uncovered background regions are encoded by applying adaptive prediction from either the neighbouring static background or a special background memory. Experimental results, obtained by computer simulations, confirm that the presented coding algorithm is a promising scheme for the application at extremely low transmission bit rates. This is shown by comparing the picture qualities obtained with the presented algorithm and a block-based hybrid-DCT scheme corresponding to H.261/RMS at 11 kbit/s.
引用
收藏
页码:228 / 235
页数:8
相关论文
共 13 条
[1]  
[Anonymous], 1989, DESCRIPTION REF MODE
[2]  
Bierling M., 1988, VIS COMMUN IMAGE P 8, P942
[3]  
GERKEN P, 1993, DEC P WORKSH MOB MUL
[4]  
GERKEN P, 1989, 2ND P INT WORKSH 64
[5]  
Gilge M., 1989, Signal Processing: Image Communication, V1, P153, DOI 10.1016/0923-5965(89)90007-6
[6]  
Hotter M., 1990, Signal Processing: Image Communication, V2, P409, DOI 10.1016/0923-5965(90)90027-F
[7]  
HOTTER M, SIGNAL PROCESSING IM
[8]  
HOTTER M, UNPUB IEEE T CIRC SY
[9]   AN INTRODUCTION TO ARITHMETIC CODING [J].
LANGDON, GG .
IBM JOURNAL OF RESEARCH AND DEVELOPMENT, 1984, 28 (02) :135-149
[10]  
MUSMANN HG, 1989, IMAGE COMMUN, V1, P117