In Defense of 3D-Label Stereo
Author
Summary, in English
It is commonly believed that higher order smoothness
should be modeled using higher order interactions. For example,
2nd order derivatives for deformable (active) contours
are represented by triple cliques. Similarly, the 2nd
order regularization methods in stereo predominantly use
MRF models with scalar (1D) disparity labels and triple
clique interactions. In this paper we advocate a largely
overlooked alternative approach to stereo where 2nd order
surface smoothness is represented by pairwise interactions
with 3D-labels, e.g. tangent planes. This general paradigm
has been criticized due to perceived computational complexity
of optimization in higher-dimensional label space.
Contrary to popular beliefs, we demonstrate that representing
2nd order surface smoothness with 3D labels leads
to simpler optimization problems with (nearly) submodular
pairwise interactions. Our theoretical and experimental results
demonstrate advantages over state-of-the-art methods
for 2nd order smoothness stereo.
should be modeled using higher order interactions. For example,
2nd order derivatives for deformable (active) contours
are represented by triple cliques. Similarly, the 2nd
order regularization methods in stereo predominantly use
MRF models with scalar (1D) disparity labels and triple
clique interactions. In this paper we advocate a largely
overlooked alternative approach to stereo where 2nd order
surface smoothness is represented by pairwise interactions
with 3D-labels, e.g. tangent planes. This general paradigm
has been criticized due to perceived computational complexity
of optimization in higher-dimensional label space.
Contrary to popular beliefs, we demonstrate that representing
2nd order surface smoothness with 3D labels leads
to simpler optimization problems with (nearly) submodular
pairwise interactions. Our theoretical and experimental results
demonstrate advantages over state-of-the-art methods
for 2nd order smoothness stereo.
Department/s
Publishing year
2013
Language
English
Pages
1730-1737
Publication/Series
Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on
Links
Document type
Conference paper
Publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
Topic
- Computer Vision and Robotics (Autonomous Systems)
- Mathematics
Conference name
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2013
Conference date
2013-06-23 - 2013-06-28
Conference place
Portland, Oregon, USA., United States
Status
Published
Research group
- Mathematical Imaging Group
ISBN/ISSN/Other
- ISSN: 1063-6919
- ISSN: 2163-6648