Abstract
In this paper, a new low complexity integer transform is proposed, which has been adopted by AVS1-P7. The proposed transform can enable AVS1-P7 to share the same quantization/dequantization table with AVS1-P2. As the bases of the proposed transform coefficients are very close, the transform normalization can be implemented only on the encoder side and the dequantization table size can be reduced compared with the transform used in H.264/MPEG-4 AVC. Along with the feature of the proposed transform, adaptive dead-zone quantization optimization for the proposed transform is studied. Experimental results show that the proposed integer transform has similar coding performance compared with the transform used in H.264/MPEG-4 AVC, and would gain near 0.1dB with the adaptive dead-zone quantization optimization.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Tran T D. The BinDCT: Fast multiplierless approximation of the DCT. IEEE Signal Processing Letters, 2000, 7(6): 141–144.
Chen Y J, Oraintara S, Tran T D et al. Multiplierless approximation of transforms with adder constraint. IEEE Signal Processing Letters, 2002, 9(11): 344–347.
Chen Y J, Oraintara S, Nguyen T Q. Video compression using integer DCT. In Proc. Int. Conf. Image Processing, Vancouver, Canada, September 10–13, 2000, pp.844–845.
Bjontegaard G. Coding improvement by using 4 × 4 blocks for motion vectors and transform. ftp://ftp3.itu.int/video-site/9712_Eib/q15c23.doc.
Wien M, Sun S J. ICT comparison for adaptive block transform. ftp://ftp3.itu.int/video-site/0101_Eib/VCEG-L12.doc.
Hallapuro A, Karczewicz M, Malvar H. Low complexity transform and quantization—Part I: Basic implementation. ftp://ftp3.itu.int/video-site/0201_Gen/JVT-B038.doc.
Hallapuro A, Karczewicz M, Malvar H. Low complexity transform and quantization—Part II: Extensions. ftp://ftp3.itu.int/video-site/0201_GenJVT-B039.doc.
Wien M. Clean-up and improved design consistency for ABT. ftp://ftp3.itu.int/jvt-site/2002_10_Geneva/JVT-E025.doc.
Malvar H S. Low-complexity length-4 transform and quantization with 16-bit arithmetic. ftp://ftp.imtc-files.org/jvt-experts/0109_San/VCEG-N44.doc.
Gordon S, Marpe D, Wiegand T. Simplified use of 8 × 8 transforms—proposal. ftp://ftp3.itu.int/jvt-site/2003_12_Waikoloa/JVT-J029.doc.
Gordon S, Marpe D, Wiegand T. Simplified use of 8 × 8 transforms—updated proposal and results. ftp://ftp3.itu.int/jvt-site/2004_03_Munich/JVT-K028.doc.
Zhang C X, Lou J, Yu L et al. The techniques of pre-scaled integer transform. In Proc. Int. Symp. Circuits and Systems, Kobe, Japan. May 23–26, 2005, pp.316–319.
Lou J, Dong J, Yu L. 8 × 8 integer transform. ftp://159.226.42.57/pulic/AVS_M1182.doc.
Wiegand T. H.26L Test Model Long-Term Number 9 (TML-9) draft0. ftp://ftp.imtc-files.org/0109_San/VCEG-N83d1.doc.
Wien M, Sun S J. ICT comparison for adaptive block transforms. ftp://ftp.imtc-files.org/jvt-experts/0109_San/VCEG-L12.doc.
Tourapis A M, Boyce J. Performance evaluation of the 8 × 8 transform vs. coefficient adaptive deadzone. ftp://ftp.imtc-files.org/jvt-experts/0403_Munich/JVT-K035.doc.
Sullivan G. Adaptive quantization encoding technique using an equal expected-value rule. ftp://ftp3.itu.int/jvt-site/2005_01_HongKong/JVT-N011.doc.
AVS FTP Site. ftp://159.226.42.57.
Author information
Authors and Affiliations
Corresponding author
Additional information
Supported by the National Natural Science Foundation of China under Grant No. 60333020, and the National Basic Research 973 Program of China under Grant No. 2001CCA03300.
Rights and permissions
About this article
Cite this article
Ma, SW., Gao, W. Low Complexity Integer Transform and Adaptive Quantization Optimization. J Comput Sci Technol 21, 354–359 (2006). https://doi.org/10.1007/s11390-006-0354-8
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/s11390-006-0354-8