Collection method and device of target area in picture and computer readable storage medium

Abstract

The invention relates to the field of image processing, and discloses a method and a device for acquiring a target area in a picture, and a computer readable storage medium. The method for obtaining the target area in the picture comprises the steps of inputting a to-be-detected picture into a detection model, and obtaining a target detection point in the to-be-detected picture; wherein the detection model comprises a backbone module and multiple stages of detection modules connected with the backbone module; the backbone module is used for acquiring feature maps of the to-be-detected picturein each preset dimension; the multi-stage detection module is used for carrying out position detection and category detection on the feature map and determining a target detection point according to adetection result of the position detection and a detection result of the category detection; and obtaining a target area from the to-be-detected picture according to the target detection point. Compared with the prior art, the method and device for obtaining the target area in the picture and the computer readable storage medium have the advantages that the target area where the to-be-recognizedtarget is located can be accurately detected, and the picture can be conveniently and accurately cut.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method and device for acquiring a target area in a picture, and a computer-readable storage medium. Background technique [0002] With the development of modern information technology in the direction of intelligence and humanization, various human-computer interaction, virtual reality, and intelligent monitoring systems have emerged one after another. Technologies such as human body pose estimation, action recognition, and behavior understanding based on computer vision play an important role in it. [0003] However, the inventors of the present invention found that the action recognition technology in the prior art usually needs to use a deep learning model, and when the deep learning model performs action recognition on a picture, if the action image of the picture is small, for example, the picture of a football match During action recognition, if the football game picture ...

AI Technical Summary

Problems solved by technology

However, due to the inability to accurately crop the area where the target to be recognized is located in the existing technology, it is necessary to perform action recognition on all the small pictures, which leads to a double increase in the amount of calculation

Images