SPLATNet: Sparse Lattice Networks for Point Cloud Processing. We present a network architecture for processing point clouds that directly operates on a collection of points represented as a sparse set of samples in a high-dimensional lattice. Naively applying convolutions on this lattice scales poorly, both in terms of memory and computational cost, as the size of the lattice increases. Instead, our network uses sparse bilateral convolutional layers as building blocks. These layers maintain efficiency by using indexing structures to apply convolutions only on occupied parts of the lattice, and allow flexible specifications of the lattice structure enabling hierarchical and spatially-aware feature learning, as well as joint 2D-3D reasoning. Both point-based and image-based representations can be easily incorporated in a network with such layers and the resulting model can be trained in an end-to-end manner. We present results on 3D segmentation tasks where our approach outperforms existing state-of-the-art techniques.
Keywords for this software
References in zbMATH (referenced in 4 articles )
Showing results 1 to 4 of 4.
- Ramasinghe, Sameera; Khan, Salman; Barnes, Nick; Gould, Stephen: Representation learning on unit ball with 3D roto-translational equivariance (2020)
- Rosu, Radu Alexandru; Quenzel, Jan; Behnke, Sven: Semi-supervised semantic mapping through label propagation with semantic texture meshes (2020)
- Zhang, Wenxiao; Long, Chengjiang; Yan, Qingan; Chow, Alix L. H.; Xiao, Chunxia: Multi-stage point completion network with critical set supervision (2020)
- Jens Behley, Martin Garbade, Andres Milioto, Jan Quenzel, Sven Behnke, Cyrill Stachniss, Juergen Gall: SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences (2019) arXiv