SPLATNet: Sparse Lattice Networks for Point Cloud Processing

CVPR 2018 Hang SuVarun JampaniDeqing SunSubhransu MajiEvangelos KalogerakisMing-Hsuan YangJan Kautz

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... (read more)

PDF Abstract

Evaluation results from the paper

Task Dataset Model Metric name Metric value Global rank Compare
Semantic Segmentation ScanNet SPLAT Net 3DIoU 0.393 # 11
3D Part Segmentation ShapeNet-Part SPLATNet 3D Class Average IoU 82.0 # 3
3D Part Segmentation ShapeNet-Part SPLATNet 3D Instance Average IoU 84.6 # 6