Training method of video optimization model and electronic equipment

Abstract

The invention belongs to the field of image processing, and provides a training method of a video optimization model and electronic equipment, and the method comprises the steps: inputting a fuzzy video into a preset video optimization model for alignment and deblurring and obtaining a first output image in an output video; determining an optical flow according to a label video, and transforming the first output image according to the optical flow to obtain a second output image; determining the difference between the first output image and the first label image and the difference between the second output image and the second label image, and adjusting the parameters of the video optimization model according to the differences until the training of the video optimization model is completed. The to-be-optimized video is subjected to alignment processing through the alignment module, the aligned image is subjected to video optimization through the trained video optimization model, the deblurred video can be obtained, meanwhile, the time domain consistency problem can be effectively optimized, and therefore the optimization efficiency of the blurred video can be effectively improved.

Description

technical field [0001] The present application relates to the field of image processing, in particular to a training method and electronic equipment for a video optimization model. Background technique [0002] With the improvement of the performance of electronic equipment such as mobile phones, tablets, and smart TVs, and the development of communication technology, video services are more and more widely used in electronic equipment. The high-quality video playback effect is conducive to users watching clearer picture content and improving the viewing experience of users. [0003] During video capture, due to camera shake, target movement, etc., the relative motion of the camera relative to the object and the background will cause motion blur; or, due to the change of the camera’s depth of field, the focus position of the camera will change and cause motion. Vague. Or, during the shooting process, due to the influence of shooting factors, post-editing factors or transco...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application discloses a video optimization model training method and electronic equipment to solve the problem that the time domain consistency problem and the motion blur problem cannot be processed at the same time when the video image is currently optimized, which is not conducive to improving the efficiency of video optimization.

Images