Image target recognition method, device, electronic device and computer storage medium

Abstract

The invention relates to an image target recognition method, device, electronic equipment and computer storage medium. The method includes: acquiring an image to be recognized, the to-be-recognized image including a target object; decomposing the target object in layers to obtain at least two layers of node; take the bottom layer as the current layer, repeat the following steps until the previous layer of the current layer is the top layer, and determine the recognition result of the target object according to the predicted recognition result of the topmost node: identify the current layer The first association relationship between each node; according to the identification result of each node of the current layer and the first association relationship, the identification result of each parent node of the upper layer of the current layer is predicted. Through the method of the present invention, the layered decomposition of the target object reduces the influence of the occluded part of the target object on the unoccluded part when identifying the target object, so that the method of the present invention can be well Adapt to target occlusion scenes.

Description

technical field [0001] The present invention relates to the technical fields of computer, image processing and machine learning, and in particular, the present invention relates to a method, apparatus, electronic device and computer storage medium for identifying an image target. Background technique [0002] With the rapid development of camera technology and computer vision technology, image instance segmentation has become a hot research direction. Image instance segmentation is the semantic segmentation of all instances of a multi-class (or single-class) object from an image. Although the current image instance segmentation methods have a good segmentation effect, for complex image scenes, the instance segmentation accuracy still needs to be improved, because the existing image instance segmentation methods have insufficient processing ability for target occlusion, and image context. The recognition ability is also limited. Therefore, for an image in which the target o...

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved by the present invention is to provide a recognition method, device, electronic equipment and computer storage medium of an image target, aiming at solving the problem that the target object cannot be accurately recognized in an occluded image

Images