Motion action judgment method based on human body recognition

The invention discloses a motion action judgment method based on human body recognition, and the method comprises the following steps: S1, capturing an image of a to-be-recognized object through AI recognition for a to-be-recognized object person; s2, calculating a motion track T1 of the skeleton key points of the to-be-recognized object by using the obtained three-dimensional position coordinates Z of the skeleton key points; s3, human body depth image information is collected through AI recognition, data processing is carried out by combining the position coordinates of the quaternion of the skeleton joints and utilizing 3D graphics, and the motion type, the motion amount and the motion action are obtained; and S4, realizing joint point capture, image data output, three-dimensional space position data output and quaternion position data output through a schematic diagram of 15 skeleton joint feature points covering the human body captured by skeleton tracking. According to the invention, the exerciser is re