jagilley/controlnet-pose – Run with an API on Replicate: The ControlNet Pose tool is pretty cool because it can create images that match the exact pose of the person in the input image! It does this by using Stable Diffusion and Controlnet to copy the weights of neural network blocks and then uses a “locked” and “trainable” copy. You can adjust different settings like the number of samples, image resolution, guidance scale, seed, eta, added prompt, negative prompt, and resolution for detection. And the best part? The predictions are usually ready in just 21 seconds!
ControlNet Pose Use Cases – Ai Tools
Modify images with humans using pose detection
Leave a Reply