A method and apparatus for video super-resolution reconstruction
A super-resolution reconstruction and resolution technology, applied in the field of image processing, can solve the problems of being unable to meet the real-time video playback requirements and consuming large equipment processing resources, so as to meet the real-time playback requirements, reduce equipment resource consumption, and improve super-resolution Effect of Resolution Reconstruction Efficiency
Active Publication Date: 2022-04-22
CHINANETCENT TECH
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, since video can be considered as a large number of continuous images, using the original image super-resolution reconstruction scheme to perform super-resolution reconstruction of video frame by frame will consume a large amount of equipment processing resources, and at the same time cannot meet the real-time playback of video need
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
[0029] In order to make the object, technical solution and advantages of the present invention clearer, the implementation manner of the present invention will be further described in detail below in conjunction with the accompanying drawings.
[0030] The embodiment of the present invention provides a method for video super-resolution reconstruction. The method is applicable to any network device with image processing function. The network device can be provided with a deep learning framework for training and generating a super-resolution reconstruction model. , and at the same time, an accelerator framework can be provided for using the existing super-resolution reconstruction model to perform image reasoning operations, thereby realizing super-resolution reconstruction processing. This embodiment will take the accelerator framework as an example of NVIDIA TensorRT framework for illustration. In terms of hardware configuration, the network device can also be equipped with CPU...
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 discloses a video super-resolution reconstruction method and device, belonging to the technical field of image processing. The method includes: generating engine files corresponding to a plurality of preset resolutions based on a convolutional neural network and an accelerator framework, wherein the engine files are used to perform super-resolution reconstruction of a video; creating multiple working threads based on a GPU device, Set the private data of each working thread; according to the resolution of the video to be reconstructed, determine the target engine file, and load the target engine file on a plurality of selected target working threads; Image reasoning is performed on the video data of the video to be reconstructed to generate super-resolution reconstructed video data. By adopting the present invention, super-resolution reconstruction of videos with different resolutions can be performed, the efficiency of super-resolution reconstruction can be improved, equipment resource consumption can be reduced, and the real-time playback requirement of videos can be met.
Description
technical field [0001] The present invention relates to the technical field of image processing, in particular to a method and device for video super-resolution reconstruction. Background technique [0002] Super-resolution reconstruction technology is a method of converting low-resolution images into high-resolution images with higher pixel density and more detailed information through a certain algorithm. With the continuous development of deep learning technology, super-resolution reconstruction of images through convolutional neural networks has become a research focus. It uses a large number of high-resolution images to construct a material library to train and generate super-resolution reconstruction models, and then use The super-resolution reconstruction model performs super-resolution reconstruction on low-resolution images. [0003] In the specific process, the high-resolution image can be degraded first to generate a low-resolution image, and the high-resolution ...
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): G06T3/40G06N3/04G06T5/00
CPCG06T3/4053G06T5/00G06T2207/20081G06T2207/20084G06T2207/10016G06N3/045
Inventor 陈祥祥郑振贵
Owner CHINANETCENT TECH