Classification of Point Cloud Scenes with Multiscale Voxel Deep Network
This work addresses point cloud classification for applications like urban planning or robotics, but it is incremental as it builds on existing benchmarks and methods.
The authors tackled the problem of classifying 3D point clouds in urban or indoor scenes by developing a new convolutional neural network that uses only point positions in multi-scale neighborhoods, achieving second place on the reduced-8 Semantic3D benchmark and beating state-of-the-art methods without regularization.
In this article we describe a new convolutional neural network (CNN) to classify 3D point clouds of urban or indoor scenes. Solutions are given to the problems encountered working on scene point clouds, and a network is described that allows for point classification using only the position of points in a multi-scale neighborhood. On the reduced-8 Semantic3D benchmark [Hackel et al., 2017], this network, ranked second, beats the state of the art of point classification methods (those not using a regularization step).