Unsigned Distance Field as an Accurate 3D Scene Representation for Neural Scene Completion

TL;DR

This paper introduces an Unsigned Distance Function (UDF) for 3D scene completion, offering a more accurate and efficient alternative to traditional Signed Distance Functions, especially for open surfaces, demonstrated on indoor and outdoor datasets.

Contribution

The paper proposes a novel UDF representation for scene completion that does not require normal computation and improves accuracy over traditional SDFs.

Findings

UDF improves scene completion accuracy on various datasets.

UDF can be computed from any point cloud without normal estimation.

Explicit geometry can be extracted from UDF on sparse grids.

Abstract

Scene Completion is the task of completing missing geometry from a partial scan of a scene. Most previous methods compute an implicit representation from range data using a Truncated Signed Distance Function (T-SDF) computed on a 3D grid as input to neural networks. The truncation decreases but does not remove the border errors introduced by the sign of SDF for open surfaces. As an alternative, we present an Unsigned Distance Function (UDF) as an input representation to scene completion neural networks. The proposed UDF is simple, and efficient as a geometry representation, and can be computed on any point cloud. In contrast to usual Signed Distance Functions, our UDF does not require normal computation. To obtain the explicit geometry, we present a method for extracting a point cloud from discretized UDF values on a sparse grid. We compare different SDFs and UDFs for the scene completion task on indoor and outdoor point clouds collected using RGB-D and LiDAR sensors and show improved completion using the proposed UDF function.