Video human body track tracking method and device and storage medium

The invention discloses a video human body track tracking method and system and a storage medium. The video human body track tracking method and system are used for tracking and recognizing a human body required in a video. The method comprises: the advantage that the number of SqueezeNet training parameters is small is utilized, effective information of the human body module is extracted by usingan RCNN (Region wish CNN) algorithm; the Adaboost algorithm is used for detecting a face image; obtaining the texture information of the face image through an LBP (Local Binding Pattern) algorithm; The method comprises the following steps: obtaining video frames of a human body through human body module detection, utilizing the advantages of a mean value distance algorithm, roughly obtaining thevideo frames of the human body through human body module detection, eliminating some non-important video images, judging whether the obtained video frames are required video frames or not through human face detection, and fina