Image segmentation method, network training method, electronic equipment and storage medium

Abstract

The invention discloses an image segmentation method, a training method of an image segmentation network, electronic equipment and a storage medium. The image segmentation method comprises the following steps: sending an input image to an image segmentation network; obtaining a first foreground target box in the input image; acquiring a first region of interest and a first region of interest feature map in the input image based on the first foreground target box; dividing the first region of interest into a plurality of grids, predicting a feature corresponding to each grid in the first region of interest based on a first region of interest feature map, and obtaining a semantic feature of each pixel point in the first region of interest; and obtaining an instance segmentation result based on the feature of each grid in the first region of interest, the information of the first foreground target box and the semantic feature of each pixel point in the first region of interest. In this way, the memory occupied by image segmentation can be reduced, and the consumed time can be shortened.

Description

technical field [0001] The present application relates to the technical field of image segmentation, in particular to an image segmentation method, a training method for an image segmentation network, electronic equipment and a storage medium. Background technique [0002] Image segmentation refers to pixel-level classification of images to segment objects (such as cars, people, trees, sky, etc.) in the image, so that subsequent processing of the target area can be achieved. Image segmentation methods can be divided into semantic segmentation, instance segmentation, panoramic segmentation and so on. Semantic segmentation can obtain the specific categories of all pixels in the image (such as cars, people, trees, sky, etc.), while instance segmentation can obtain the specific categories of each target pixel in the image (such as cars, people, etc.) , to segment different targets of the same category. Panoramic segmentation combines semantic segmentation and instance segmentat...

AI Technical Summary

Problems solved by technology

[0004] This application provides an image segmentation method, an image segmentation network training method, electronic equipment and storage media, which can solve the problems that the existing instance segmentation methods are not fine enough, need to occupy a large amount of memory, and need to consume a lot of time.

Images