Keywords: multi-camera/multi-hypothesis/multi-target segmentation and tracking, background subtraction, visual surveillance, human detection/tracking and activity analysis, image/video quality measurement.

	Multi-camera (multi-view), multi-hypothesis, multy-target segmentation and tracking: An efficient automatic technique for tracking multiple targets from multiple cameras (or views) becomes valuable for law enforcement, military and commercial applications.  A multi-view multi-hypothesis approach to segmenting and tracking multiple persons on a ground plane is proposed. The tracking state space is the set of ground points of the people being tracked. During tracking, several iterations of segmentation are performed using information from human appearance models and ground plane homography. Kyungnam Kim and Larry S. Davis, "Multi-Camera Tracking and Segmentation of Occluded People on Ground Plane using Search-Guided Particle Filtering", European Conference on Computer Vision (ECCV), LNCS, 2006.
	Background Modeling and Subtraction: Robust methods for segmenting moving foreground objects from the background scene.  This is a sub-project of VACE-ADVICE (Activity Detection by Video Content Extraction) of ARDA (Advanced Research and Development Activity). (2000-now) K. Kim, T. H. Chalidabhongse, D. Harwood and L. Davis, "Real-time Foreground-Background Segmentation using Codebook Model", Real-time Imaging, Volume 11, Issue 3, Pages 167-256, June 2005. Kyungnam Kim, D. Harwood and Larry S. Davis, "Background Updating for Visual Surveillance", International Symposium on Visual Computing (ISVC), LNCS 3804, 2005. K. Kim, T. H. Chalidabhongse, D. Harwood and L. Davis, "Background Modeling and Subtraction by Codebook Construction" IEEE International Conference on Image Processing (ICIP) 2004.
	PDR (Perturbation Detection Rate) analysis: We introduce a performance evaluation methodology called Perturbation Detection Rate (PDR) analysis, for measuring performance of background subtraction (BGS) algorithms. It has some advantages over the commonly used Receiver Operation Characteristics (ROC) analysis. Specifically, it does not require foreground targets or knowledge of foreground distributions. It measures the sensitivity of a BGS algorithm in detecting low contrast targets against background as a function of contrast, also depending on how well the model captures mixed (moving) background events. (2001-now) T. H. Chalidabhongse, K. Kim, D. Harwood and L. Davis, "A Perturbation Method for Evaluating Background Subtraction Algorithms", Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance (VS-PETS 2003), Nice, France, Oct. 11-12, 2003.
	I designed a tool to assess no-reference objective quality of image or video using local statistics of pixels over space and time.  It reports several quality measures – fine structure quality, clipping, entropy, etc. Ÿ   Q1 – noise: due to sensor sensitivity, electronic transmission, illumination fluctuation, camera vibration, etc Ÿ   Q2 – contrast: affected by camera optics, resolution, etc Ÿ   Q3 – color information: how well color values are distributed over the intensity range (entropy). Ÿ   Q4 – clipping: pixels brighter or darker than  the representation bounds (ex, [0,255]) are clipped. K. Kim and L. Davis, "A Fine-Structure Image/Video Quality using Local Statistics" IEEE International Conference on Image Processing (ICIP) 2004.
	NASA Crew Performance Analyzer: This project is to track NASA crew in the space shuttle and to analyze their motions, postures, etc.  The ultimate goal of this project is to find out their cognitive states and to help designing shuttle interfaces, etc in ergonomics-view. (2001-2002) Sample video

• Background Subtraction Results on ARDA Sample-Event-Videos (Jul. 2003)

• Application materials for summer internship 2003.

• ColorBrightness_Apr10_2003(page1, page2, page3, page4), ColorBrightness_Apr15_2003 (page1, page2, page3, page4)

• Spatial Temporal Filtering - support filter (page1 page2 page3 page4) (Jul. 2003)

• CB retraining (Oct. 2003)

• Codeword variance update (Feb. 2004)

• I am a coordinator of the computer vision study group 'KG-VISA'

UMD-BGS: a foreground-background segmentation (background subtraction) program developed by Computer Vision Lab, University of Maryland.  It accepts several image formats, video file and live video.

“This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder."

Journal Papers

• K. Kim, T. H. Chalidabhongse, D. Harwood and L. Davis, "Real-time Foreground-Background Segmentation using Codebook Model", Real-time Imaging, Volume 11, Issue 3, Pages 167-256, June 2005.

Conference papers

• Kyungnam Kim and Larry S. Davis, "Multi-Camera Tracking and Segmentation of Occluded People on Ground Plane using Search-Guided Particle Filtering", European Conference on Computer Vision (ECCV), LNCS, 2006.

• Kyungnam Kim, D. Harwood and Larry S. Davis, "Background Updating for Visual Surveillance", International Symposium on Visual Computing (ISVC), LNCS 3804, 2005.

• K. Kim, T. H. Chalidabhongse, D. Harwood and L. Davis, "Background Modeling and Subtraction by Codebook Construction" IEEE International Conference on Image Processing (ICIP) 2004.

• K. Kim and L. Davis, "A Fine-Structure Image/Video Quality using Local Statistics" IEEE International Conference on Image Processing (ICIP) 2004.

• T. H. Chalidabhongse, K. Kim, D. Harwood and L. Davis, "A Perturbation Method for Evaluating Background Subtraction Algorithms", Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance (VS-PETS 2003), Nice, France, Oct. 11-12, 2003. 

• Kyungnam Kim, Jong-Gook Ko, SeungHo Choi, JinYoung Kim and KiJung Kim, "An Experimental Multimodal System Integrating Speech Recognition, Lip Reading, and Face and Gaze Tracking" International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC) 2000, Pusan, Korea, Jul. 2000.

• Jong-Gook Ko, Kyungnam Kim, SeungHo Choi, JinYoung Kim, KiJung Kim,and JungNyo Kim, "Facial Feature Tracking and Head Orientation-based Gaze Tracking”, International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC) 2000, Pusan, Korea, Jul. 2000.

• Kyungnam Kim, R. S. Ramakrishna, “Human Computer Interface using Vision-based eye-gaze tracking”, Human Computer Interaction ’99, Korea, 2000.

• Kyung-Nam Kim, R. S. Ramakrishna, "Vision-Based Eye-Gaze Tracking for Human Computer Interface", IEEE International Conf. on Systems, Man, and Cybernetics, Tokyo, Japan, Oct.1999. [956KB]

• Jong-Gook Ko, Kyung-Nam Kim, R.S. Ramakrishna, "Facial Feature Tracking for Eye-Head Controlled Human Computer Interface", IEEE, TENCON'99, Cheju, Korea, Sept. 1999. [174KB]

• S.-H. Kim, Y.-J. Suh, K.-N. Kim, J.-G. Ko, R. S. Ramakrishna, "Design of Parallel Processing System for Face Tracking”, Proceeding of '98 Korea Information Science Society Fall Conference, 1998. [249KB]

• K.-N. Kim, R. S. Ramakrishna, T.-S. Choi, "New Parallel Vision Environment in Heterogeneous Networked Computing”, Proceeding of SPIE, Vol. 3452, SPIE 43rd Annual Meeting, San Diego, CA, 1998. [347KB]

• Other 5 publications

Invited Papers

• T. H. Chalidabhongse, K. Kim, D. Harwood and L. Davis, "A Perturbation Method for Evaluating Background Subtraction Algorithms", Performance Metrics for Intelligent Systems Workshop 2003, NIST, Gaithersburg, MD, Sep. 16-18, 2003.

Under preparation

• "Spatial/temporal support filter", "Color models for detection and tracking algorithms", "Variance update and normalization (same FA, consistent encoding/detection rates)", "Parameter estimation for background subtraction algorithms", "Neighborhood template matching", "24/7 video surveillance systems", "New spatial filter", also see KenPublish.doc.

Patent

• Kyung-Nam Kim, "Methods of eye-gaze tracking based on computer vision", Korea Patent, 2001

Personal memo (for prospective computer vision students)

• Kyungnam Kim, "How to PhD-research" 

Copyright ⓒ 1998, Kyungnam Kim. All rights reserved.  

Last updated: Sep. 29, 2006
Kyungnam Kim