Current Research

Detection and Location of People in Video Images Using Adaptive Fusion of Color and Edge Information

Sumer Jabri, Zoran Duric, Harry Wechsler, and Azriel Rosenfeld.

Abstract

A new method of finding people in video images is presented. Detection is based on a novel background modeling and subtraction approach which uses both color and edge information. We introduce confidence maps - gray-scale images whose intensity is a function of our confidence that a pixel has changed - to fuse intermediate results and to represent the results of background subtraction. The latter is used to delineate a person's body by guiding contour collection to segment the person from the background. The method is tolerant to scene clutter, slow illumination changes, and camera noise, and runs in near real time on a standard platform.

Results

Indoor, using Sony 3CCD progressive digital camera. The segmented human sequences show the results of detection and delineation, with the delineating contour in white.

Input Sequence Segmented Human
(1,498K) (1,444K)
(1,131K) (1,132K)
(625K) (647K)
(1,013K) (915K)
(954K) (1,380K)

 

Outdoor, using Sony CCD TR500 Handycam. As in the previous sequences, the segmented human sequences show the results of detection and delineation, with the delineating contour in white.

Input Sequence Segmented Human
(1,262K) (1,072K)
(395K) (178K)
(521K) (253K)

 

System Interface

(95K)

 


Tracking Groups of People

Stephen McKenna, Sumer Jabri, Zoran Duric, Harry Wechsler, and Azriel Rosenfeld.

Abstract

A computer vision system for tracking multiple people in relatively unconstrained environments is described. Tracking is performed at three levels of abstraction: regions, people and groups. A novel, adaptive background subtraction method that combines color and gradient information is used to cope with shadows and unreliable color cues. People are tracked through mutual occlusions as they form groups and separate from one another. Strong use is made of color information to disambiguate occlusions and to provide qualitative estimates of depth ordering and position during occlusion. Some simple interactions with objects can also be detected. The system is tested using both indoor and outdoor sequences. It is robust and should provide a useful mechanism for bootstrapping and reinitialization of tracking using more specific but less robust human models.

Results

Outdoor, using Sony CCD TR500 Handycam. As in the previous sequences, the segmented human sequences show the results of detection and delineation, with the delineating contour in white.

Input Sequence Segmented Human
(1,262K) (863K)
(395K) (134K)
(521K) (256K)

 


[Home | Publications | Resume | Projects]

Computer Vision and Neural Networks Laboratory

Department of Computer Science

George Mason University