LHM is a Large Animatable Human Reconstruction Model that enables the creation of detailed digital human models from a single image in seconds, featuring a core inference pipeline, model deployment, and motion extraction capabilities.
