MergeNet: Explicit Mesh Reconstruction from Sparse Point Clouds via Edge Prediction

TL;DR

MergeNet is a new explicit mesh reconstruction method from sparse point clouds that predicts edge connections to improve efficiency and accuracy, outperforming existing explicit techniques especially with sparse data.

Contribution

The paper introduces MergeNet, a novel approach that predicts local edge connectivity for mesh reconstruction, addressing issues of face selection and sparsity.

Findings

MergeNet outperforms state-of-the-art explicit methods on synthetic and real datasets.

It effectively filters edges based on predicted surface distances.

The method improves mesh quality from sparse point clouds.

Abstract

This paper introduces a novel method for reconstructing meshes from sparse point clouds by predicting edge connection. Existing implicit methods usually produce superior smooth and watertight meshes due to the isosurface extraction algorithms~(e.g., Marching Cubes). However, these methods become memory and computationally intensive with increasing resolution. Explicit methods are more efficient by directly forming the face from points. Nevertheless, the challenge of selecting appropriate faces from enormous candidates often leads to undesirable faces and holes. Moreover, the reconstruction performance of both approaches tends to degrade when the point cloud gets sparse. To this end, we propose MEsh Reconstruction via edGE~(MergeNet), which converts mesh reconstruction into local connectivity prediction problems. Specifically, MergeNet learns to extract the features of candidate edges and regress their distances to the underlying surface. Consequently, the predicted distance is utilized to filter out edges that lay on surfaces. Finally, the meshes are reconstructed by refining the triangulations formed by these edges. Extensive experiments on synthetic and real-scanned datasets demonstrate the superiority of MergeNet to SoTA explicit methods.