LatticeNet: Fast Spatio-Temporal Point Cloud Segmentation Using Permutohedral Lattices
This addresses the problem of efficient and accurate 3D segmentation for applications like robotics or autonomous driving, with incremental improvements in speed and memory.
The paper tackled 3D semantic segmentation from raw point clouds by proposing LatticeNet, which uses a permutohedral lattice for fast convolutions and low memory, achieving state-of-the-art performance on multiple datasets.
Deep convolutional neural networks (CNNs) have shown outstanding performance in the task of semantically segmenting images. Applying the same methods on 3D data still poses challenges due to the heavy memory requirements and the lack of structured data. Here, we propose LatticeNet, a novel approach for 3D semantic segmentation, which takes raw point clouds as input. A PointNet describes the local geometry which we embed into a sparse permutohedral lattice. The lattice allows for fast convolutions while keeping a low memory footprint. Further, we introduce DeformSlice, a novel learned data-dependent interpolation for projecting lattice features back onto the point cloud. We present results of 3D segmentation on multiple datasets where our method achieves state-of-the-art performance. We also extend and evaluate our network for instance and dynamic object segmentation.