3D Menagerie: Modeling the 3D shape and pose of animals

Silvia Zuffi, Angjoo Kanazawa, David Jacobs, Michael J. Black

Published 2016-11-23Version 1

There has been significant prior work on learning realistic, articulated, 3D statistical shape models of the human body. In contrast, there are few such models for animals, despite their many applications. The main challenge is that animals are much less cooperative subjects than humans. The best human body models are learned from thousands of 3D scans of people in specific poses, which is infeasible with live animals. Consequently, here we extend a state-of-the-art articulated 3D human body model to animals and learn it from a limited set of 3D scans of toy figurines in arbitrary poses. We employ a novel part-based shape model to compute an initial registration to the scans. We then normalize their pose, learn a statistical shape model, and refine the alignments and the model together. In this way, we accurately align animal scans from different quadruped families with very different shapes and poses. With the alignment to a common template we learn a shape space representing animals including lions, cats, dogs, horses, cows and hippos. Animal shapes can be sampled from the model, posed, animated, and fitted to data. In particular, we demonstrate the generalization of the model by fitting it to images of real animals, and show that it captures realistic animal shapes, even for new species not seen in training. We make our model available for research, enabling the extension of methods for human shape and pose estimation to animals.