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.

翻译：有效模拟变形,现有方法代表3D对象,使用多边形模模模,使用皮革技术进行变形。本文介绍了神经分解形状近似值(NASA),这是一个替代框架,它能有效表示以外表为条件的神经指示功能的变形物体。使用美国航天局的占用测试是直截了当的,绕过模具的复杂性和水密问题。我们展示了美国航天局对3D跟踪应用的有效性,并讨论了其他可能的扩展。