Deep Parametric Continuous Convolutional Neural Networks

Standard convolutional neural networks assume a grid structured input is available and exploit discrete convolutions as their fundamental building blocks. This limits their applicability to many real-world applications. In this paper we propose Parametric Continuous Convolution, a new learnable oper...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 2589 - 2597
Main Authors Wang, Shenlong, Suo, Simon, Ma, Wei-Chiu, Pokrovsky, Andrei, Urtasun, Raquel
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2018
Subjects
Computer architecture
Convolution
Kernel
Neural networks
Three-dimensional displays
Online AccessGet full text

Cover

More Information
Summary:Standard convolutional neural networks assume a grid structured input is available and exploit discrete convolutions as their fundamental building blocks. This limits their applicability to many real-world applications. In this paper we propose Parametric Continuous Convolution, a new learnable operator that operates over non-grid structured data. The key idea is to exploit parameterized kernel functions that span the full continuous vector space. This generalization allows us to learn over arbitrary data structures as long as their support relationship is computable. Our experiments show significant improvement over the state-of-the-art in point cloud segmentation of indoor and outdoor scenes, and lidar motion estimation of driving scenes.
ISSN:1063-6919
DOI:10.1109/CVPR.2018.00274