Image processing method and device, computer equipment and storage medium

Abstract

The invention relates to an image processing method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining an original image and a target generation size; expanding the original image according to the ratio of the height to the width of the target generation size to obtain an expanded image; generating a mask image according to the expansion area and the original area of the expansion image, wherein the mask image and the expansion image have the same dimension; and inputting the mask image, the target generation size and the original image into an image generation network, outputting a generated image, the size of the generated image being the same as the target generation size, and the image generation network being a sub-network of the target adversarial network. Through the trained image generation network, the original image features are extracted and reconstructed to obtain the image adaptive to the display device, so that the generated image has more complete image information, the image is played, and the user experience can be improved.

Description

Technical field [0001] This application relates to the field of computer technology, and in particular to an image processing method, device, computer equipment and storage medium. Background technique [0002] As an important carrier of entertainment media, video carries important information. With the rise of self-media and UGC, more and more users use their own shooting equipment to record their lives and upload them to the Internet. [0003] When a shooting device shoots a video, a large number of the videos are shot on the vertical screen, and the user uploads the video to the Internet. When the video is played, it is often played on the horizontal screen. At present, the video is turned from vertical to horizontal, and the saliency detection method is usually used to detect the main body of the video, and the main target is cut out from the video, and the main part of the video is cut out, which causes information loss and affects the look and feel. Summary of the invention...

AI Technical Summary

Problems solved by technology

At present, the video is converted from vertical to horizontal, and the saliency detection method is usually used to detect the main body of the video, and the main object is cut out from the video, and the main part of the video is cut, resulting in information loss and affecting the perception.

Images