A super-resolution reconstruction method, device, equipment and storage medium

Abstract

Embodiments of the present application disclose an artificial intelligence-based super-resolution reconstruction method, device, equipment and storage medium. When a video file needs to be super-resolution reconstructed, a sequence of video frames to be processed at the first resolution is obtained. , the video frame sequence to be processed is the video frame sequence of the video file. For each to-be-processed video frame in the to-be-processed video frame sequence, perform resolution reconstruction on the to-be-processed video frame according to the resolution reconstruction model to obtain an initial reconstructed video frame of a second resolution; the second resolution is higher than the first resolution. The contour area in the initial reconstructed video frame is determined through the residual separation method, and the contour area is enhanced to obtain the target reconstructed video frame, which makes the high-frequency details clearer and obtains optimal image quality results that meet the actual online application scenarios. In this way, generating a second resolution reconstructed video frame sequence based on the target reconstructed video frame can provide users with a cleaner, clearer, natural and comfortable video experience.

Description

technical field [0001] The present application relates to the field of image processing, in particular to an artificial intelligence-based super-resolution reconstruction method, device, device and storage medium. Background technique [0002] Image super-resolution refers to the technology of reconstructing a high-resolution image by processing one or more low-resolution images with complementary information, which is widely used in various media data (such as video or image) quality, such as online video viewing, medical imaging, video surveillance, remote sensing imaging and other fields. [0003] A super-resolution reconstruction method based on deep learning has been proposed in related technologies, but the generalization effect of the super-resolution reconstruction method in related technologies is poor. Facing real scenes online, the results obtained may not be the results with the best picture quality, often Satisfactory super-resolution effects cannot be obtained...

AI Technical Summary

Problems solved by technology

[0003] A super-resolution reconstruction method based on deep learning has been proposed in related technologies, but the generalization effect of the super-resolution reconstruction method in related technologies is poor. Facing real scenes online, the obtained results may not be the results with the best picture quality, often Unable to achieve satisfactory super-resolution effect

Images