Image repair method, device, equipment and storage medium

Abstract

The present application provides an image repair method, device, equipment and storage medium. The method includes: marking the area of ​​interest in the target image to obtain the marked area image, and determining the boundary information of the marked area image; segmenting the marked area image according to the boundary information to obtain the first image, and segmenting the target image to obtain the second image; fusing the first image and the second image to obtain a fused image; inputting the fused image and the first image into an image repair model to determine the target repair image, the image repair model is used to determine the target repair image for repairing the first image; repairing the first image according to the target repair image. The present application realizes automatic identification of central holes and boundary holes that cannot be identified by the naked eye during the image annotation process, and automatically repairs the central holes and boundary holes to improve the image annotation efficiency and annotation quality.

Images