Jump to Content

NASA: Neural Articulated Shape Approximation

Boyang Deng
JP Lewis
Timothy Jeruzalski
Gerard Pons-Moll
Geoffrey Hinton
Mohammad Norouzi
European Conference on Computer Vision (ECCV) (2020)


Efficient representation of articulated objects such as human bodies is an important problem in computer vision and graphics. To efficiently simulate deformation, existing approaches represent 3D objects using polygonal meshes and deform them using skinning techniques. This paper introduces neural articulated shape approximation (NASA), an alternative framework that enables efficient representation of articulated deformable objects using neural indicator functions that are conditioned on pose. Occupancy testing using NASA is straightforward, circumventing the complexity of meshes and the issue of water-tightness. We demonstrate the effectiveness of NASA for 3D tracking applications, and discuss other potential extensions.

Research Areas