We synthesize comprehensive-physique human pictures starting from a supplied human pose with two devoted methods. 1) With a few texts describing the shapes of clothes, the specified human pose is initially translated into a human parsing map. 2) The ultimate human picture is then generated by supplying the system with https://www.humanizingaitext.com/