A neural network-based video coding system and method
A neural network and video coding technology, applied in the field of video coding systems, can solve the problems of low compression efficiency, avoid block effects, reduce the tediousness of manually designing encoders, and flexibly combine coding systems
Active Publication Date: 2021-09-17
NORTHWESTERN POLYTECHNICAL UNIV
View PDF11 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Patent (3) discards the traditional coding framework and completely adopts a neural network-based coding and decoding method, so the compression efficiency is low
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0043] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0044] Such as figure 1 As shown, the present invention provides a neural network-based video coding system, which effectively combines the neural network with the traditional video coding framework, which can not only utilize the existing fine design in the traditional framework, but also play the role of the neural network in data The role of the compressed field. In the coding process described in the present invention, each step can freely adopt the traditional method or the neural network method, or adaptively select one of the coding methods according to the coding rate, and the whole system is flexible and effective.
[0045] To achieve the above object, the present invention adopts the following technical solutions:
[0046] A neural network-based video coding system, including time-domain down-sampling module, intra-frame codec module, inter-...
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
Login to View More Abstract
The invention provides a neural network-based video coding system, which includes a time-domain down-sampling module, an intra-frame codec module, an inter-frame codec module, and a time-domain up-sampling module. The video coding system effectively combines the neural network with the traditional video coding framework, which can not only utilize the existing fine design in the traditional framework, but also play the role of the neural network in the field of data compression. The combination of the coding system in the present invention is relatively flexible, and each coding component can be based on a traditional method or a neural network method. The present invention also provides a video encoding method using a neural network-based video encoding system, which can reduce the complexity of manually designing a large number of encoders and greatly improve encoding efficiency.
Description
technical field [0001] The invention belongs to the field of signal and information processing, and in particular relates to a video encoding system and a division method. Background technique [0002] Since the late 1980s, the Video Coding Expert Group of the ITU Telecommunication Standardization Sector and the Dynamic Image Expert Group of the International Organization for Standardization / International Electrotechnical Commission have jointly promulgated a series of international standards on video coding, the main standard of which is H. 264 / AVC standard, H.265 / HEVC standard and H.266 / VVC standard. [0003] These standards are based on a hybrid coding framework of predictive coding and transform coding, where each module in the framework (intra prediction, inter prediction, transform, quantization, inverse transform, inverse quantization, entropy coding, loop filtering) requires A large number of fine artificial design [0004] In recent years, Artificial Neural Networ...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): H04N19/132H04N19/587H04N19/59H04N19/176H04N19/91H04N19/593H04N19/503H04N19/86H04N19/119H04N19/577
CPCH04N19/119H04N19/132H04N19/176H04N19/503H04N19/577H04N19/587H04N19/59H04N19/593H04N19/86H04N19/91
Inventor 万帅王铭泽梅少辉巩浩
Owner NORTHWESTERN POLYTECHNICAL UNIV