|
Stephen Krotosky, Ph.D. |
|
|
Graduate Student Researcher Phone:
(858) 822-0002 |
|
|
Research Interests |
|
|
I recently successfully defended my thesis to earn my Ph.D. in Electrical Engineering from the University of California, San Diego. My dissertation, On Using Multiperspective Color and Thermal Infrared Videos to Detect People: Issues, Computational Framework, Algorithms and Comparative Analysis, is the culmination of my six years spent studying in San Diego. I am currently employed as an Algorithm Development Engineer at SAIC.
My Ph.D. work is based on combining color and thermal infrared video for looking at people in a scene. I am interested in the problem of registering color and infrared features in scenes with parallax, as well as exploring how multimodality can increase the rate and robustness of person detection.
In past research, I have investigated
the use of stereo cameras to detect occupant 3D head locations inside of an
automobile for the purpose of intelligent airbag deployment. In addition to
the development of the head detection algorithms, a framework for ground
truth analysis and real-time trials has also been laid out. Computer Vision and Robotics
Research Laboratory (CVRR) |
|
|
Publications |
|
|
On Color, Infrared and Multimodal
Stereo Approaches to Pedestrian Detection Krotosky,
S.J.; Trivedi, M.M; IEEE Trans. On Intelligent Transportation Systems,
IN PRESS. A Comparison of Color and Infrared
Stereo Approaches to Pedestrian Detection Krotosky,
S.J.; Trivedi, M.M; Proc. IEEE Intelligent Vehicles Symposium, Jun.
2007. (pdf) Mutual Information Based
Registration of Multimodal Stereo Videos for Person Tracking Registration
of Multimodal Imagery with Occluding Objects using Mutual Information Registration
of Multimodal Stereo Images using Disparity Voting from Correspondence
Windows Multimodal
Stereo Image Registration for Pedestrian Detection Multimodal
Image Registration for Person Detection: Analysis and Review Real-Time
Stereo-Based Head Detection using Size, Shape and Disparity Constraints Face
Detection and Head Tracking using Stereo and Thermal Infrared Cameras for
'Smart' Airbags: A Comparative Analysis Occupant
Posture Analysis with Stereo and Thermal Infrared Video: Algorithms and
Experimental Evaluation Occupant
Posture Analysis using Reflectance and Stereo Images for "Smart"
Airbag Deployment Detection
and identification of sardine eggs at sea using a machine vision system |
|
|
Last Modified
October 31, 2007,
copyright © 2007 Stephen Krotosky |
|