Image coding and decoding method and device
An image coding and entropy coding technology, which is applied in image communication, digital video signal modification, electrical components, etc., can solve the problems of high computing power consumption for image coding and decoding, and large amount of deep neural network parameters.
Active Publication Date: 2022-05-27
HUAWEI TECH CO LTD +1
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] However, since the prediction of the probability of the feature map needs to consider the local context information, the prediction of the probability of each element needs to wait for the decoding of the neighboring elements of the element Complete, so that the system coupling of the deep neural network is strong, and the parameters of the deep neural network are large, which leads to a large computational power consumption of image encoding and decoding
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
[0099] The term "and / or" in this article is only an association relationship to describe the associated objects, indicating that there can be three kinds of relationships, for example, A and / or B, it can mean that A exists alone, A and B exist at the same time, and A and B exist independently. B these three cases.
[0100] The terms "first" and "second" in the description and claims of the embodiments of the present application are used to distinguish different objects, rather than to describe a specific order of the objects. For example, the first feature map, the second feature map, the third feature map, etc. are used to distinguish different feature maps, rather than to describe a specific order of the feature maps; the first reconstruction map, the second reconstruction map, and the third reconstruction map The graphs, etc. are used to distinguish different reconstruction graphs, not to describe the specific order of the reconstructed graphs; for another example, the firs...
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 an image coding and decoding method and device, relates to the technical field of image coding and decoding, and aims to reduce parameters of a neural network and reduce computing power consumption of image coding and decoding. The method comprises the following steps: performing first decorrelation transformation on a target image by using a first analysis network to obtain a first feature map of the target image; performing second decorrelation transformation on the first feature map by using a second analysis network to obtain a second feature map of the target image; carrying out entropy coding on the second feature map based on a second probability histogram of the second feature map to obtain a second code stream of the target image; obtaining a second reconstruction image of the target image based on a second probability histogram and a second code stream of the second feature map; taking the second reconstruction image as prior information, and performing probability prediction on the first feature image by using a first probability prediction network to obtain a first probability histogram; and performing entropy coding on the first feature map based on the first probability histogram of the first feature map to obtain a first code stream of the target image.
Description
technical field [0001] The embodiments of the present application relate to the technical field of image encoding and decoding, and in particular, to an image encoding and decoding method and apparatus. Background technique [0002] With the development of neural network technology, the image encoding and decoding (ie image compression) method based on deep learning (deep neural network) has also emerged, and the performance of image encoding and decoding based on deep learning has surpassed the traditional hybrid encoding and decoding framework. codec performance. [0003] Taking image encoding as an example, in one implementation, the deep neural network includes a transformation network and a probability prediction network. When encoding an image based on the deep neural network, the image to be encoded is transformed by the transformation network to obtain the characteristics of the image to be encoded. After the map, entropy coding is performed on each element of the f...
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 IPC IPC(8): H04N19/13H04N19/61
CPCH04N19/13H04N19/61
Inventor 刘家瑛胡越予王晶
Owner HUAWEI TECH CO LTD