IIT ECE Ph.D. Candidate

Maral Mesma Khosroshahi

Ph.D. candidate

Phone
Email mmesmakh@hawk.iit.edu
Web www.iit.edu/~mmesmakh
Resume Resume

Education

  • B. S., EE, Sharif University of Technology, Tehran, Iran, 2010
  • M. S., ECE, Illinois Institute of Technology, Chicago, 2012.

Research

Maral joined the department of Electrical and Computer Engineering at Illinois Institute of Technology in Fall 2010. She received her Master degree in May 2012 and is currently a second year Ph.D. candidate and graduate research assistant at the Multimedia Communication Lab under the supervision of Prof. Joohee Kim. Her research interests include computer vision, pattern recognition, machine learning, object detection and object tracking.

Selected Publications

  1. M. Mesmakhosroshahi and J. Kim, "ROI Reduction for Fast Pedestrian Detection," submitted to the Visual Communications and Image Processing (VCIP), 2015.

  2. M. Zarshenas, M. Mesmakhosroshahi and J. Kim, "Fast Depth Estimation using Spatio-temporal Prediction for Stereo-based Pedestrian Detection," submitted to the Visual Communications and Image Processing (VCIP), 2015.

  3. M. Mesmakhosroshahi, C. Zhang and J. Kim, "Stereo based ROI Generation and Tracking for Pedestrian Detection," submitted to the Image and Vision Computing.

  4. M. Mesmakhosroshahi, K-H Chung, Y. Lee, J. Kim, "Depth Gradient Based Region of Interest Generation for Pedestrian Detection," In the International SoC Design Conference (ISOCC), 2014.

  5. J. Kim and M. Mesmakhosroshahi, "Stereo-based Region of Interest Generation for Real-time Pedestrian Detection," In Peer-to-Peer Networking and Applications, 2013. <pdf> <Link>

  6. M. Mesmakhosroshahi, J. Kim, Y. Lee, and J-B. Kim, "Stereo based Region of Interest Generation for Pedestrian Detection in Driver Assistance Systems," In Proceedings of the International Conference on Image Processing (ICIP), pp. 3386-3389, 2013. <pdf> <Link>

  7. M. Mesmakhosroshahi and J. Kim, "Improving spatio-temporal feature extraction techniques and their applications in action classification," VCIP 2012. <pdf> <Link>

PATENTS AND INVENTION DISCLOSURES

  • M. Mesmakhosroshahi, M. Loghman and J. Kim, "A system for detecting pedestrians by fusing color and depth information," U.S. Patent # 62/030,994.
  • M. Loghman, M. Mesmakhosroshahi and J. Kim, "Multi-resolution Census transform based dense disparity estimation for advanced driver assistance systems," U.S. Patent # 62/037,987.