The human body model method ignores the complex conditions inside the human body, and uses simple two-dimensional graphics or line diagrams such as ellipses, rectangles, and trapezoids to describe the skeleton or various parts of the human body, as shown in the figure below, and then matches the detected targets with these models. So as to determine whether it is a human target. This kind of method is dedicated to the intuitive expression of human body shape, which is easy to analyze and understand the human body posture and movement. However, when the background is more complex, there may be more targets that conform to the characteristics of the simple model, and the recognition accuracy will drop significantly. If a model with a complex structure is used, due to the various poses of the human target, when the human target is self-occluded or blocked by other targets, it may not have the characteristics of the complex model used at all, resulting in the correlation between the human target and the model. Failed, a large number of targets were missed. In addition, since the human body is a non-rigid object, it will be very difficult to locate the starting point and junction of the corresponding model. In addition, when multiple objects appear in the scene at the same time, there are many choices for the matching direction of the model, so it is difficult to determine whether there are only human objects in the image. Or a combination of various goals. In general, this method has a narrow scope of application and low reliability.