Tangent Convolutions for Dense Prediction in 3D

We present an approach to semantic scene analysis using deep convolutional networks. Our approach is based on tangent convolutions - a new construction for convolutional networks on 3D data. In contrast to volumetric approaches, our method operates directly on surface geometry. Crucially, the constr...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 3887 - 3896
Main Authors Tatarchenko, Maxim, Park, Jaesik, Koltun, Vladlen, Zhou, Qian-Yi
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2018
Subjects
Convolution
Geometry
Kernel
Semantics
Shape
Three-dimensional displays
Two dimensional displays
Online AccessGet full text

Cover

More Information
Summary:We present an approach to semantic scene analysis using deep convolutional networks. Our approach is based on tangent convolutions - a new construction for convolutional networks on 3D data. In contrast to volumetric approaches, our method operates directly on surface geometry. Crucially, the construction is applicable to unstructured point clouds and other noisy real-world data. We show that tangent convolutions can be evaluated efficiently on large-scale point clouds with millions of points. Using tangent convolutions, we design a deep fully-convolutional network for semantic segmentation of 3D point clouds, and apply it to challenging real-world datasets of indoor and outdoor 3D environments. Experimental results show that the presented approach outperforms other recent deep network constructions in detailed analysis of large 3D scenes.
ISSN:1063-6919
DOI:10.1109/CVPR.2018.00409