Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces

TL;DR

Neural-Pull introduces a neural network-based method that learns signed distance functions by iteratively pulling query points onto surfaces, resulting in more accurate 3D surface reconstructions from point clouds.

Contribution

It presents a simple, differentiable pulling operation that improves the accuracy and flexibility of learning signed distance functions for surface reconstruction.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Achieves higher quality surface reconstructions from point clouds.

Effective for both 3D surface and single image reconstruction.

Abstract

Reconstructing continuous surfaces from 3D point clouds is a fundamental operation in 3D geometry processing. Several recent state-of-the-art methods address this problem using neural networks to learn signed distance functions (SDFs). In this paper, we introduce \textit{Neural-Pull}, a new approach that is simple and leads to high quality SDFs. Specifically, we train a neural network to pull query 3D locations to their closest points on the surface using the predicted signed distance values and the gradient at the query locations, both of which are computed by the network itself. The pulling operation moves each query location with a stride given by the distance predicted by the network. Based on the sign of the distance, this may move the query location along or against the direction of the gradient of the SDF. This is a differentiable operation that allows us to update the signed distance value and the gradient simultaneously during training. Our outperforming results under widely used benchmarks demonstrate that we can learn SDFs more accurately and flexibly for surface reconstruction and single image reconstruction than the state-of-the-art methods.