Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Optimisation of a quantisation matrix for image and video coding

A technology of quantization matrix and image sequence, applied in the field of quantization matrix, it can solve the problem that the method is not optimal

Inactive Publication Date: 2006-06-28
THOMSON LICENSING SA
View PDF0 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these perceptually optimal methods are not optimal for high bitrate encoding

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Optimisation of a quantisation matrix for image and video coding
  • Optimisation of a quantisation matrix for image and video coding
  • Optimisation of a quantisation matrix for image and video coding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0066] Several adaptive calculation methods of quantization matrices for both intra blocks and inter blocks will be described below. These methods can be used in all discrete cosine transform (DCT) based image or video coding standards, such as JPEG, MPEG-2 and MPEG-4 H.264 FRExt, and provide flexibility to the quantization process to improve subjective or objective quality, and even adjust the bit rate. For high definition (HD) video coding, 8*8 resizing performs better than 4*4 resizing. Therefore, in the following description, for intra blocks and inter blocks, 4*4 transformation is disabled, and quantization matrices both have a size of 8*8, if not stated otherwise.

[0067] figure 1 The average distribution of the amplitude levels (ie histogram) of the 64 DCT coefficients of all intra-frame 8*8 blocks in the HDTV sequence Kung_fu is shown. Each small image corresponds to the DCT position. The horizontal coordinate is the quantized amplitude (level), and the vertical c...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

Disclosed is a method and apparatus for generating a quantization matrix for encoding an image or image sequence, wherein the effect of greatly reducing the data rate in video encoding is achieved by quantizing the transformed frequency coefficients or components of a pixel block, so that only Fewer amplitude levels are encoded, and a portion of quantized amplitudes becomes 0 and does not need to be encoded as quantized amplitudes. The quantization matrix generation of the present invention starts with a default quantization matrix and thus results in the perceptually optimal quantization matrix for a given sequence of images. In the first pass, candidate quantization matrices are iteratively constructed for a given sequence of images by simultaneously increasing the scaling values ​​for some coefficient positions and decreasing the scaling values ​​for some other coefficient positions. In the second pass, the resulting quantization matrix is ​​applied to the image sequence for re-encoding.

Description

technical field [0001] The invention relates to a method and a device for adaptively generating quantization matrices usable for encoding an image or image sequence. Background technique [0002] A significant data rate reduction effect in video coding can be achieved by quantizing the (transformed) frequency coefficients or components of a pixel block, so that thereafter only fewer amplitude levels need to be encoded, and a part of the quantized amplitude becomes 0 and does not need to be encoded as a quantized magnitude. Many transform-based video coding standards use default quantization matrices to achieve better subjective video coding / decoding quality, such as ISO / IEC (International Organization for Standardization / International Electrotechnical Commission) 13818-2 (MPEG-2 Video). The 'quantization matrix' assigns smaller scaling values ​​(i.e. with larger divisors) to certain frequencies of the block if the associated horizontal and / or vertical frequencies are consid...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): H04N7/26
CPCH04N19/172H04N19/134H04N19/61H04N19/60H04N19/126H04N19/154H04N19/177H04N19/174H04N19/192H04N19/124
Inventor 陈颖翟杰夫
Owner THOMSON LICENSING SA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products