Learning the 3D Fauna of the Web

arXiv:2401.02400v2 Announce Type: replace
Abstract: Learning 3D models of all animals on the Earth requires massively scaling up existing solutions. With this ultimate goal in mind, we develop 3D-Fauna, an approach that learns a pan-category deformable 3D animal model for more than 100 animal species jointly. One crucial bottleneck of modeling animals is the limited availability of training data, which we overcome by simply learning from 2D Internet images. We show that prior category-specific attempts fail to generalize to rare …

3d models abstract animals arxiv cs.cv earth fauna mind modeling scaling scaling up solutions species type web