DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds
This addresses the problem of learning from 3D point-cloud data for applications in computer vision and robotics, offering a novel method for handling directional information on surfaces.
The paper tackled the challenge of constructing anisotropic convolution layers for 3D point clouds by introducing DeltaConv, which uses geometric operators from vector calculus to enable anisotropic filtering, resulting in matching or improving state-of-the-art performance on benchmarks and speeding up training and inference.
Learning from 3D point-cloud data has rapidly gained momentum, motivated by the success of deep learning on images and the increased availability of 3D~data. In this paper, we aim to construct anisotropic convolution layers that work directly on the surface derived from a point cloud. This is challenging because of the lack of a global coordinate system for tangential directions on surfaces. We introduce DeltaConv, a convolution layer that combines geometric operators from vector calculus to enable the construction of anisotropic filters on point clouds. Because these operators are defined on scalar- and vector-fields, we separate the network into a scalar- and a vector-stream, which are connected by the operators. The vector stream enables the network to explicitly represent, evaluate, and process directional information. Our convolutions are robust and simple to implement and match or improve on state-of-the-art approaches on several benchmarks, while also speeding up training and inference.