Efficient large size transforms for high-performance video coding

  title={Efficient large size transforms for high-performance video coding},
  author={Rajan L. Joshi and Yuriy A. Reznik and Marta Karczewicz},
  booktitle={Optical Engineering + Applications},
This paper describes design of transforms for extended block sizes for video coding. The proposed transforms are orthogonal integer transforms, based on a simple recursive factorization structure, and allow very compact and efficient implementations. We discuss techniques used for finding integer and scale factors in these transforms, and describe our final design. We evaluate efficiency of our proposed transforms in VCEG's H.265/JMKTA framework, and show that they achieve nearly identical… 
Integer Cosine Transforms for High-Efficiency Image and Video Coding
One-norm order-8 and -16 integer transforms are proposed and fast algorithms are developed that implement these transforms with a low computational complexity that is less by a factor of 3–5 than the complexity of well-known algorithms.
Simple High-Order Integer Cosine Transform for High-Resolution Video Coding
A one-norm simple order-32 integer transform is constructed and its fast algorithms of low computational complexity are developed that require only integer operations and whose computational complexity is 4.9 times less than that of well-known algorithms.
A Method to Construct a Large-Size Simple Integer Cosine Transform for High-Resolution Image and Video Coding
A matrix method is considered to construct simple order-32 integer cosine step transform and its fast algorithms are developed whose computational complexity is 4.3 times smaller than that of well-known algorithms.
Performance analysis of Integer DCT of different block sizes
This project discusses how the use of larger transforms, especially in high resolution videos, can provide higher coding gain and proposes transforms of sizes larger than 4x4 or 8x8, especially 16x16 and 32x32 are proposed because of their increased applicability to the de-correlation of high resolution video signals.
High-Order Integer Cosine Transform: Construction Method and Fast Algorithms for High-Resolution Image and Video Coding
The one-norm order-32 integer transform is constructed and its fast algorithms of low computing complexity are developed and this transform is not only very close to the discrete cosine transform but also has excellent coding performance.
A Hybrid Video Coder Based on Extended Macroblock Sizes, Improved Interpolation, and Flexible Motion Representation
The proposed video codec follows a hybrid coding approach based on temporal prediction, followed by transform, quantization, and entropy coding of the residual, and was among the best-performing CfP proposals.
Design of high-performance fixed-point transforms using the common factor method
  • A. Hinds
  • Computer Science, Engineering
    Optical Engineering + Applications
  • 2010
The Common Factor method is reviewed as a method to arrive at fixed-point implementations that are optimized in terms of complexity and precision for such high performance transforms such as the 16x16 DCT.
A hybrid video codec based on extended block sizes, recursive integer transforms, improved interpolation, and flexible motion representation
Simulation results are presented for a variety of bit rates, resolutions and coding configurations to demonstrate the high compression efficiency achieved by the proposed video codec at moderate level of encoding and decoding complexity.
Point DCT VLSI Architecture for Emerging HEVC Standard
The proposed architecture relies on the decomposition of the DCT matrices into sparse submatrices in order to reduce the multiplications and is partially folded to save area and to get speed for large video sequences sizes.
Video Description Length Guided Constant Quality Video Coding with Bitrate Constraint
Experimental results show that the proposed coding strategy with saved computation could achieve better or similar RD performance than other coding strategies.


Design of fast transforms for high-resolution image and video coding
In the construction and analysis, an array of known techniques are utilized, including Heideman's mapping between DCT and DFT, Winograd short length DFT modules, prime-factor and common-factor algorithms, and also a new factorization scheme for even-sized scaled transforms are offered.
Efficient implementation of a class of MDCT/IMDCT filterbanks for speech and audio coding applications
  • R. Chivukula, Y. Reznik
  • Computer Science
    2008 IEEE International Conference on Acoustics, Speech and Signal Processing
  • 2008
An efficient algorithm for implementing MDCT/IMDCT of lengths N = 5 ldr 2m (m ges 2) is presented and it is shown that such a modified window is piece-wise symmetric, and can be stored using N/2 words.
HD Photo: a new image coding technology for digital photography
This paper introduces the HD Photo coding technology developed by Microsoft Corporation. The storage format for this technology is now under consideration in the ITU-T/ISO/IEC JPEG committee as a
Standardization of IDCT approximation behavior for video compression: the history and the new MPEG-C parts 1 and 2 standards
  • G. Sullivan
  • Computer Science
    SPIE Optical Engineering + Applications
  • 2007
The history of international standardization activities and specifications relating to the precision of inverse discrete cosine transform approximations used in video compression designs and the development of the new MPEG-C Part 1 and Part 2 standards are presented.
Efficient fixed-point approximations of the 8×8 inverse discrete cosine transform
This paper describes fixed-point design methodologies and several resulting implementations of the Inverse Discrete Cosine Transform (IDCT) contributed by the authors to MPEG's work on defining the
A Fast Computational Algorithm for the Discrete Cosine Transform
A Fast Discrete Cosine Transform algorithm has been developed which provides a factor of six improvement in computational complexity when compared to conventional Discrete Cosine Transform algorithms
Improved precision of fixed-point algorithms by means of common factors
A general technique for improving the precision of fixed-point implementations of signal processing algorithms relying on the use of "common factors", which are applied to groups of real constants in the algorithms turning them into quantities that can be more accurately approximated by dyadic rational numbers.