B. Hua, M. Tran, и S. Yeung. (2017)cite arxiv:1712.05245Comment: 10 pages, 6 figures, 10 tables. Paper accepted to CVPR 2018.
Аннотация
Deep learning with 3D data such as reconstructed point clouds and CAD models
has received great research interests recently. However, the capability of
using point clouds with convolutional neural network has been so far not fully
explored. In this paper, we present a convolutional neural network for semantic
segmentation and object recognition with 3D point clouds. At the core of our
network is pointwise convolution, a new convolution operator that can be
applied at each point of a point cloud. Our fully convolutional network design,
while being surprisingly simple to implement, can yield competitive accuracy in
both semantic segmentation and object recognition task.
%0 Generic
%1 hua2017pointwise
%A Hua, Binh-Son
%A Tran, Minh-Khoi
%A Yeung, Sai-Kit
%D 2017
%K 2018 arxiv cnn computer-vision cvpr paper point-cloud
%T Pointwise Convolutional Neural Networks
%U http://arxiv.org/abs/1712.05245
%X Deep learning with 3D data such as reconstructed point clouds and CAD models
has received great research interests recently. However, the capability of
using point clouds with convolutional neural network has been so far not fully
explored. In this paper, we present a convolutional neural network for semantic
segmentation and object recognition with 3D point clouds. At the core of our
network is pointwise convolution, a new convolution operator that can be
applied at each point of a point cloud. Our fully convolutional network design,
while being surprisingly simple to implement, can yield competitive accuracy in
both semantic segmentation and object recognition task.
@misc{hua2017pointwise,
abstract = {Deep learning with 3D data such as reconstructed point clouds and CAD models
has received great research interests recently. However, the capability of
using point clouds with convolutional neural network has been so far not fully
explored. In this paper, we present a convolutional neural network for semantic
segmentation and object recognition with 3D point clouds. At the core of our
network is pointwise convolution, a new convolution operator that can be
applied at each point of a point cloud. Our fully convolutional network design,
while being surprisingly simple to implement, can yield competitive accuracy in
both semantic segmentation and object recognition task.},
added-at = {2018-07-20T09:16:41.000+0200},
author = {Hua, Binh-Son and Tran, Minh-Khoi and Yeung, Sai-Kit},
biburl = {https://www.bibsonomy.org/bibtex/262867296c54a0bbffd80b49f7735f2ad/analyst},
description = {[1712.05245] Pointwise Convolutional Neural Networks},
interhash = {3a163ab1e8a0406cc46df672eb459187},
intrahash = {62867296c54a0bbffd80b49f7735f2ad},
keywords = {2018 arxiv cnn computer-vision cvpr paper point-cloud},
note = {cite arxiv:1712.05245Comment: 10 pages, 6 figures, 10 tables. Paper accepted to CVPR 2018},
timestamp = {2018-07-20T09:16:41.000+0200},
title = {Pointwise Convolutional Neural Networks},
url = {http://arxiv.org/abs/1712.05245},
year = 2017
}