FPNN: Field Probing Neural Networks for 3D Data

by   Yangyan Li, et al.

Building discriminative representations for 3D data has been an important task in computer graphics and computer vision research. Convolutional Neural Networks (CNNs) have shown to operate on 2D images with great success for a variety of tasks. Lifting convolution operators to 3D (3DCNNs) seems like a plausible and promising next step. Unfortunately, the computational complexity of 3D CNNs grows cubically with respect to voxel resolution. Moreover, since most 3D geometry representations are boundary based, occupied regions do not increase proportionately with the size of the discretization, resulting in wasted computation. In this work, we represent 3D spaces as volumetric fields, and propose a novel design that employs field probing filters to efficiently extract features from them. Each field probing filter is a set of probing points --- sensors that perceive the space. Our learning algorithm optimizes not only the weights associated with the probing points, but also their locations, which deforms the shape of the probing filters and adaptively distributes them in 3D space. The optimized probing points sense the 3D space "intelligently", rather than operating blindly over the entire domain. We show that field probing is significantly more efficient than 3DCNNs, while providing state-of-the-art performance, on classification tasks for 3D object recognition benchmark datasets.


page 1

page 2

page 3


A Convolutional Neural Network into graph space

Convolutional neural networks (CNNs), in a few decades, have outperforme...

Learning to Prune Filters in Convolutional Neural Networks

Many state-of-the-art computer vision algorithms use large scale convolu...

Spatially-Adaptive Filter Units for Compact and Efficient Deep Neural Networks

Convolutional neural networks excel in a number of computer vision tasks...

Teaching Compositionality to CNNs

Convolutional neural networks (CNNs) have shown great success in compute...

RotDCF: Decomposition of Convolutional Filters for Rotation-Equivariant Deep Networks

Explicit encoding of group actions in deep features makes it possible fo...

Convolutional Neural Networks on 3D Surfaces Using Parallel Frames

We extend Convolutional Neural Networks (CNNs) on flat and regular domai...

ZerNet: Convolutional Neural Networks on Arbitrary Surfaces via Zernike Local Tangent Space Estimation

The research community has observed a massive success of convolutional n...

Please sign up or login with your details

Forgot password? Click here to reset