C. Ackerman, und L. Itti. IEEE Transactions on Robotics, 21 (2):
247-251(April 2005)
Zusammenfassung
We introduce a method for rapidly classifying visual scenes, globally
along a small number of navigationally relevant dimensions: depth
of scene, presence of obstacles, path vs. non-path, and orientation
of path. We show that the algorithm reliably classifies scenes in
terms of these high-level features, based on global or coarsely localized
spectral analysis analogous to early-stage biological vision. We
use this analysis to implement a real-time visual navigational system
on a mobile robot, trained online by a human operator. We demonstrate
successful training and subsequent autonomous path following for
two different out-door environments, a running track and a concrete
trail. Our success with this technique suggests a general applicability
to autonomous robot navigation in a variety of environments.
%0 Journal Article
%1 Ackerman_Itti05tro
%A Ackerman, C.
%A Itti, L.
%D 2005
%J IEEE Transactions on Robotics
%K Fourier a autonomous following gist navigation of path robot scene transform vision |
%N 2
%P 247-251
%T Robot Steering With Spectral Image Information
%V 21
%X We introduce a method for rapidly classifying visual scenes, globally
along a small number of navigationally relevant dimensions: depth
of scene, presence of obstacles, path vs. non-path, and orientation
of path. We show that the algorithm reliably classifies scenes in
terms of these high-level features, based on global or coarsely localized
spectral analysis analogous to early-stage biological vision. We
use this analysis to implement a real-time visual navigational system
on a mobile robot, trained online by a human operator. We demonstrate
successful training and subsequent autonomous path following for
two different out-door environments, a running track and a concrete
trail. Our success with this technique suggests a general applicability
to autonomous robot navigation in a variety of environments.
@article{Ackerman_Itti05tro,
abstract = {We introduce a method for rapidly classifying visual scenes, globally
along a small number of navigationally relevant dimensions: depth
of scene, presence of obstacles, path vs. non-path, and orientation
of path. We show that the algorithm reliably classifies scenes in
terms of these high-level features, based on global or coarsely localized
spectral analysis analogous to early-stage biological vision. We
use this analysis to implement a real-time visual navigational system
on a mobile robot, trained online by a human operator. We demonstrate
successful training and subsequent autonomous path following for
two different out-door environments, a running track and a concrete
trail. Our success with this technique suggests a general applicability
to autonomous robot navigation in a variety of environments.},
added-at = {2007-12-04T05:18:34.000+0100},
author = {Ackerman, C. and Itti, L.},
biburl = {https://www.bibsonomy.org/bibtex/2195dcb19b7ce366bd3cd22d0b5945780/lm77},
description = {Dissertation},
file = {http://ilab.usc.edu/publications/doc/Ackerman_Itti05tro.pdf},
interhash = {9f9c94c4898fae4445dfcd3bdba6625a},
intrahash = {195dcb19b7ce366bd3cd22d0b5945780},
journal = {IEEE Transactions on Robotics},
keywords = {Fourier a autonomous following gist navigation of path robot scene transform vision |},
month = Apr,
number = 2,
pages = {247-251},
timestamp = {2007-12-04T05:18:35.000+0100},
title = {Robot Steering With Spectral Image Information},
type = {bb | cv | sc},
volume = 21,
year = 2005
}