Detecting Vanishing Points using Global Image Context in a Non-Manhattan
World
M. Zhai, S. Workman, and N. Jacobs. (2016)cite arxiv:1608.05684Comment: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016.
Abstract
We propose a novel method for detecting horizontal vanishing points and the
zenith vanishing point in man-made environments. The dominant trend in existing
methods is to first find candidate vanishing points, then remove outliers by
enforcing mutual orthogonality. Our method reverses this process: we propose a
set of horizon line candidates and score each based on the vanishing points it
contains. A key element of our approach is the use of global image context,
extracted with a deep convolutional network, to constrain the set of candidates
under consideration. Our method does not make a Manhattan-world assumption and
can operate effectively on scenes with only a single horizontal vanishing
point. We evaluate our approach on three benchmark datasets and achieve
state-of-the-art performance on each. In addition, our approach is
significantly faster than the previous best method.
Description
[1608.05684] Detecting Vanishing Points using Global Image Context in a Non-Manhattan World
%0 Generic
%1 zhai2016detecting
%A Zhai, Menghua
%A Workman, Scott
%A Jacobs, Nathan
%D 2016
%K cnn deep detection estimation horizon learning line network neural point vanishing
%T Detecting Vanishing Points using Global Image Context in a Non-Manhattan
World
%U http://arxiv.org/abs/1608.05684
%X We propose a novel method for detecting horizontal vanishing points and the
zenith vanishing point in man-made environments. The dominant trend in existing
methods is to first find candidate vanishing points, then remove outliers by
enforcing mutual orthogonality. Our method reverses this process: we propose a
set of horizon line candidates and score each based on the vanishing points it
contains. A key element of our approach is the use of global image context,
extracted with a deep convolutional network, to constrain the set of candidates
under consideration. Our method does not make a Manhattan-world assumption and
can operate effectively on scenes with only a single horizontal vanishing
point. We evaluate our approach on three benchmark datasets and achieve
state-of-the-art performance on each. In addition, our approach is
significantly faster than the previous best method.
@misc{zhai2016detecting,
abstract = {We propose a novel method for detecting horizontal vanishing points and the
zenith vanishing point in man-made environments. The dominant trend in existing
methods is to first find candidate vanishing points, then remove outliers by
enforcing mutual orthogonality. Our method reverses this process: we propose a
set of horizon line candidates and score each based on the vanishing points it
contains. A key element of our approach is the use of global image context,
extracted with a deep convolutional network, to constrain the set of candidates
under consideration. Our method does not make a Manhattan-world assumption and
can operate effectively on scenes with only a single horizontal vanishing
point. We evaluate our approach on three benchmark datasets and achieve
state-of-the-art performance on each. In addition, our approach is
significantly faster than the previous best method.},
added-at = {2019-02-25T17:22:24.000+0100},
author = {Zhai, Menghua and Workman, Scott and Jacobs, Nathan},
biburl = {https://www.bibsonomy.org/bibtex/27b84aab52a65586e70394ea89751a88e/kluger},
description = {[1608.05684] Detecting Vanishing Points using Global Image Context in a Non-Manhattan World},
interhash = {eafcd9dc39bd79110d8b7d9917373489},
intrahash = {7b84aab52a65586e70394ea89751a88e},
keywords = {cnn deep detection estimation horizon learning line network neural point vanishing},
note = {cite arxiv:1608.05684Comment: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016},
timestamp = {2019-02-25T17:22:24.000+0100},
title = {Detecting Vanishing Points using Global Image Context in a Non-Manhattan
World},
url = {http://arxiv.org/abs/1608.05684},
year = 2016
}