TY - GEN
T1 - A learning approach to fixating on 3D targets with active cameras
AU - Srinivasa, Narayan
AU - Ahuja, Narendra
N1 - Funding Information:
The authors wish to express their appreciation for the many research grants from theNTU Hospital (NTUH, 92-M007), National Health Research Institute (DOH88-HR-816), the National Center of High-Perfor¬ mance Computing, the National Science Council (NSC89-2314-B-002-461-M08, NSC89-2511-S-002-025-X3, NSC90-2213-E-002-063, NSC91-2213-E-002-082-), and the Ministry of Education, Taiwan.
Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 1997.
PY - 1997
Y1 - 1997
N2 - Fixation of an active camera pair on a given target requires that the pan and tilt angles of the cameras must be set to bring the target to image centers. However, the calibration needed to achieve a specific configuration of real cameras involves tedious estimation of a number of imaging parameters. Fortunately, this excercise is not essential for fixationif images are acquired and used as feedback during the fixation process to continuously direct the cameras to the target. This paper defines a direct mapping from the changes in the direction of target motion in the image plane to changes in camera angles necessary to reduce the disparity between image center and the image plane target location. The mapping captures camera calibration, as well as other effects such as deviations from the assumed imaging model which are difficult to characterize and capture in calibration. The mapping is formulated as a task in nonlinear function approximation and learnt from real data. For computational efficiency, learning is done at multiple resolutions and using a PROBART network. Experimental results are presented using an active vision system.
AB - Fixation of an active camera pair on a given target requires that the pan and tilt angles of the cameras must be set to bring the target to image centers. However, the calibration needed to achieve a specific configuration of real cameras involves tedious estimation of a number of imaging parameters. Fortunately, this excercise is not essential for fixationif images are acquired and used as feedback during the fixation process to continuously direct the cameras to the target. This paper defines a direct mapping from the changes in the direction of target motion in the image plane to changes in camera angles necessary to reduce the disparity between image center and the image plane target location. The mapping captures camera calibration, as well as other effects such as deviations from the assumed imaging model which are difficult to characterize and capture in calibration. The mapping is formulated as a task in nonlinear function approximation and learnt from real data. For computational efficiency, learning is done at multiple resolutions and using a PROBART network. Experimental results are presented using an active vision system.
UR - http://www.scopus.com/inward/record.url?scp=84957366011&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84957366011&partnerID=8YFLogxK
U2 - 10.1007/3-540-63930-6_175
DO - 10.1007/3-540-63930-6_175
M3 - Conference contribution
AN - SCOPUS:84957366011
SN - 3540639306
SN - 9783540639305
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 623
EP - 631
BT - Computer Vision - ACCV 1998 - 3rd Asian Conference on Computer Vision, Proceedings
A2 - Chin, Roland
A2 - Pong, Ting-Chuen
PB - Springer
T2 - 3rd Asian Conference on Computer Vision, ACCV 1998
Y2 - 8 January 1998 through 10 January 1998
ER -