Farm3D: Learning Articulated 3D Animals by Distilling 2D Diffusion

arXiv:2304.10535v2 Announce Type: replace
Abstract: We present Farm3D, a method for learning category-specific 3D reconstructors for articulated objects, relying solely on "free" virtual supervision from a pre-trained 2D diffusion-based image generator. Recent approaches can learn a monocular network that predicts the 3D shape, albedo, illumination, and viewpoint of any object occurrence, given a collection of single-view images of an object category. However, these approaches heavily rely on manually curated clean training data, which are expensive to obtain. We propose a …

abstract animals arxiv cs.cv diffusion free generator image image generator learn network object objects supervision type virtual