3D Neural Edge Reconstruction

TL;DR

This paper introduces EMAP, a novel neural method for 3D edge reconstruction that captures both lines and curves from multi-view edge maps, improving detail and accuracy in 3D modeling.

Contribution

EMAP is the first approach to encode 3D edge distance and direction using UDFs from multi-view edge maps, enabling robust extraction of parametric edges.

Findings

Achieves superior 3D edge reconstruction accuracy.

Enhances neural surface reconstruction with detailed edge information.

Robustly extracts parametric 3D edges from inferred edge points.

Abstract

Real-world objects and environments are predominantly composed of edge features, including straight lines and curves. Such edges are crucial elements for various applications, such as CAD modeling, surface meshing, lane mapping, etc. However, existing traditional methods only prioritize lines over curves for simplicity in geometric modeling. To this end, we introduce EMAP, a new method for learning 3D edge representations with a focus on both lines and curves. Our method implicitly encodes 3D edge distance and direction in Unsigned Distance Functions (UDF) from multi-view edge maps. On top of this neural representation, we propose an edge extraction algorithm that robustly abstracts parametric 3D edges from the inferred edge points and their directions. Comprehensive evaluations demonstrate that our method achieves better 3D edge reconstruction on multiple challenging datasets. We further show that our learned UDF field enhances neural surface reconstruction by capturing more details.