Submanifold Sparse Convolutional Networks

5 Jun 2017  ·  Benjamin Graham, Laurens van der Maaten ·

Convolutional network are the de-facto standard for analysing spatio-temporal data such as images, videos, 3D shapes, etc. Whilst some of this data is naturally dense (for instance, photos), many other data sources are inherently sparse... Examples include pen-strokes forming on a piece of paper, or (colored) 3D point clouds that were obtained using a LiDAR scanner or RGB-D camera. Standard "dense" implementations of convolutional networks are very inefficient when applied on such sparse data. We introduce a sparse convolutional operation tailored to processing sparse data that differs from prior work on sparse convolutional networks in that it operates strictly on submanifolds, rather than "dilating" the observation with every layer in the network. Our empirical analysis of the resulting submanifold sparse convolutional networks shows that they perform on par with state-of-the-art methods whilst requiring substantially less computation. read more

PDF Abstract


  Add Datasets introduced or used in this paper

Results from the Paper

Ranked #18 on 3D Part Segmentation on ShapeNet-Part (Instance Average IoU metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
3D Part Segmentation ShapeNet-Part SSCN Instance Average IoU 86.0 # 18