ControlNet Pose
The ControlNet Pose tool can replicate the pose of a figure in an input image. The weights of neural network blocks are copied into a “locked” and “trainable” duplicate using Stable Diffusion and Controlnet. Sample size, picture resolution, guidance scale, seed, eta, additional prompt, negative prompt, and detection resolution are all user-definable. The average time for the forecasts to finish is 21 seconds.
