GraphFit: Learning Multi-scale Graph-Convolutional Representation for Point Cloud Normal Estimation
This work addresses the problem of accurate normal estimation for 3D point clouds, which is crucial for applications like computer graphics and robotics, but it appears incremental as it builds on existing graph-based methods with specific enhancements.
The authors tackled the problem of normal estimation for unstructured 3D point clouds, which is challenging due to noise, nonuniform density, and sharp edges, by proposing GraphFit, a method that uses graph convolutional representation and a multi-scale architecture to achieve state-of-the-art accuracy and robustness on benchmark datasets.
We propose a precise and efficient normal estimation method that can deal with noise and nonuniform density for unstructured 3D point clouds. Unlike existing approaches that directly take patches and ignore the local neighborhood relationships, which make them susceptible to challenging regions such as sharp edges, we propose to learn graph convolutional feature representation for normal estimation, which emphasizes more local neighborhood geometry and effectively encodes intrinsic relationships. Additionally, we design a novel adaptive module based on the attention mechanism to integrate point features with their neighboring features, hence further enhancing the robustness of the proposed normal estimator against point density variations. To make it more distinguishable, we introduce a multi-scale architecture in the graph block to learn richer geometric features. Our method outperforms competitors with the state-of-the-art accuracy on various benchmark datasets, and is quite robust against noise, outliers, as well as the density variations.