Computer Vision tools for ComfyUI
A ComfyUI workflow that computes a variety of analyses of media specifically featuring people, including (among others): segmentation of the body from the background; monocular depth estimation; normal map estimation; and pose estimation of the body, face, and hands.
A ComfyUI workflow that computes similar analyses of media featuring quadruped animals:
This set of ComfyUI workflows provide heatmaps that show which parts of an image correspond to specific words in a prompt. This can be even be used for adjectives like "angry" and "bald".
