GeoUDF: Surface Reconstruction from 3D Point Clouds via Geometry-guided Distance Representation
This addresses the long-standing problem of reconstructing surfaces from sparse point clouds for applications in computer graphics and vision, representing an incremental improvement with novel components.
The paper tackles surface reconstruction from sparse 3D point clouds by proposing GeoUDF, a learning-based method that uses geometry-guided distance representation and achieves significant advantages in accuracy, efficiency, and generality over state-of-the-art methods, as demonstrated through extensive experiments.
We present a learning-based method, namely GeoUDF,to tackle the long-standing and challenging problem of reconstructing a discrete surface from a sparse point cloud.To be specific, we propose a geometry-guided learning method for UDF and its gradient estimation that explicitly formulates the unsigned distance of a query point as the learnable affine averaging of its distances to the tangent planes of neighboring points on the surface. Besides,we model the local geometric structure of the input point clouds by explicitly learning a quadratic polynomial for each point. This not only facilitates upsampling the input sparse point cloud but also naturally induces unoriented normal, which further augments UDF estimation. Finally, to extract triangle meshes from the predicted UDF we propose a customized edge-based marching cube module. We conduct extensive experiments and ablation studies to demonstrate the significant advantages of our method over state-of-the-art methods in terms of reconstruction accuracy, efficiency, and generality. The source code is publicly available at https://github.com/rsy6318/GeoUDF.