Object Registration in Neural Fields

TL;DR

This paper analyzes the Reg-NF neural field registration method for 6-DoF object pose estimation in robotics, demonstrating its ability to improve scene representation and enable scene editing through object substitution.

Contribution

It provides an expanded analysis of neural field registration for robotics, highlighting its applications in object pose estimation and scene manipulation.

Findings

Effective 6-DoF pose estimation of known objects.

Enhanced scene representation with neural fields.

Ability to generate new scenes via object substitution.

Abstract

Neural fields provide a continuous scene representation of 3D geometry and appearance in a way which has great promise for robotics applications. One functionality that unlocks unique use-cases for neural fields in robotics is object 6-DoF registration. In this paper, we provide an expanded analysis of the recent Reg-NF neural field registration method and its use-cases within a robotics context. We showcase the scenario of determining the 6-DoF pose of known objects within a scene using scene and object neural field models. We show how this may be used to better represent objects within imperfectly modelled scenes and generate new scenes by substituting object neural field models into the scene.