A method and device for detecting head key points in human body posture recognition

Abstract

The invention provides a method and a device for detecting head key points in human body posture recognition. The method comprises the following steps: providing an image to be processed; Performing target detection processing on the to-be-processed image to obtain one or more first head detection frames; Judging the head posture of each human body in the to-be-processed image; When the head posture of the human body is the front face or the back face, the midpoint of each edge of the first head detection frame serves as four first head key points; When the head posture of the human body is aleft side surface, taking the middle point of the right longitudinal edge, the left lower vertex and the middle point of the upper transverse edge of the first head detection frame as three first headkey points; And when the head posture of the human body is the right side surface, taking the middle point of the left longitudinal edge, the right lower vertex and the middle point of the upper transverse edge of the first head detection frame as three first head key points. According to the invention, the accuracy of head key point detection in human body posture recognition can be improved.

Description

technical field [0001] The present invention relates to the field of computer vision technology, in particular to a method for detecting key points of the head in human body posture recognition, a detection device for key points of the head in human body posture recognition, electronic equipment and a computer-readable storage medium. Background technique [0002] In recent years, computer vision technology based on artificial intelligence has developed rapidly. As an important direction, human gesture recognition has great application prospects in many fields such as robot interaction, games, animation, behavior recognition, and security monitoring. [0003] Human body gesture recognition technology in the prior art mainly includes two categories: [0004] One is a top-down recognition method, that is, first locate the approximate position of the human body, and then perform specific recognition of the posture. The most commonly used method is to first use the target detec...

AI Technical Summary

Problems solved by technology

[0006] However, both of the above two types of methods have the problem of low recognition accuracy: in the first type of top-down method, the individual is checked first, and then the gesture is recognized. The steps are separated from each other and have no internal connection, which makes gesture recognition heavily dependent on The results of the previous step detection and positioning; the second type of bottom-up method first detects all key points of the human body, and then connects these key points through a clustering algorithm to form a complete individual. The key points are very unsatisfactory when the human body is relatively close. In many cases, the key points of one person will be connected to another human body, and this method cannot find body parts from global features.

Images