The ControlNet Pose tool is utilized to produce images with identical poses as the individual in the input image. It employs Stable Diffusion and Controlnet to duplicate the weights of neural network blocks into a “locked” and “trainable” duplicate. The user has the ability to specify the quantity of samples- image resolution- guidance scale- seed- eta- added prompt- negative prompt- and detection resolution. Typically- the predictions are finalized within a timeframe of 21 seconds.