[Show abstract][Hide abstract] ABSTRACT: The combination of subtraction stereo with shadow detection we propose improves people tracking in stereoscopic environments. Subtraction stereo is a stereo matching method which is fast and robust for the correspondence problem – one of the most seri-ous issues in computer vision – restricting the search range of matching to foreground regions. Shadow de-tection gives adequate foreground regions of tracked people by removing cast shadows. This leads to ac-curate three-dimensional measurement of positions in stereoscopic environment tracking. By focusing on dis-parity images obtained by subtraction stereo, we can detect people easily based on standard labeling. Ob-jects can also be measured directly in size by subtrac-tion stereo without geometric information about envi-ronments for tracking. This is important for installing the tracking system easily. To track multiple passing people, we use the extended Kalman filter to address the occlusion problem usually encountered in crowded environments. The proposed method is verified by ex-periments using unknown stereoscopic environments.
[Show abstract][Hide abstract] ABSTRACT: In this paper, we propose a method for tracking groups of people using
three-dimensional (3D) feature points obtained with use of the
Kanade-Lucas-Tomasi feature tracker (KLT) method and a stereo camera
system called “Subtraction stereo”. The tracking system
using subtraction stereo, which focuses its stereo matching algorithm to
foreground regions obtained by background subtraction, is realized using
Kalman filter based tracker. The effectiveness of the proposed method is
verified using 3D scenes of people walking, which are difficult to
[Show abstract][Hide abstract] ABSTRACT: In this paper, we propose a fast and stable human detection based on "subtraction stereo" which can measure distance information of foreground regions. Scanning an input image by detection windows is controlled in their window sizes and number using the distance information obtained from subtraction stereo. This control can skip a large number of detection windows and leads to reduce the computational time and false detection for fast and stable human detection. Additionally, we propose two-step boosting as a new training way of classifier with whole and upper human body models. Experimental results show that the proposal is faster and less false detection than the method described in the reference .
IEEE International Conference on Robotics and Automation, ICRA 2011, Shanghai, China, 9-13 May 2011; 01/2011
[Show abstract][Hide abstract] ABSTRACT: In this paper, we propose a method for human tracking using a stereo camera system called "Subtraction Stereo" and color information. The tracking system using the subtraction stereo, which focuses its stereo matching algorithm to regions obtained by background subtraction, is realized using Kalman filter. To make the tracking system more robust, the new method also uses color information as another distinctive information of person. The effectiveness of the proposed method is verified in the scene which is difficult to realize without color information.
[Show abstract][Hide abstract] ABSTRACT: This paper discusses a method to detect and measure pedestrians in urban environment using ¿subtraction stereo.¿ Subtraction stereo is a stereo vision method that calculates distance only for moving regions to make the stereo matching robust and fast. Pedestrians are detected based on their three dimensional features, which are position, height, width, and so on, obtained from a range image by standard labeling. The number of pedestrians in groups is estimated from their area on a range image. In urban environments, the effectiveness of this method is verified by experiments using a stereo camera which is commercially available and implemented with subtraction stereo.
[Show abstract][Hide abstract] ABSTRACT: In this paper, estimation of camera's extrinsic parameters in measurement of pedestrians using "subtraction stereo" is discussed. Subtraction stereo is a stereo vision method that focuses on the movement of objects to make a stereo camera robust and produces range images for moving regions. Features of pedestrians such as 3D position, height and width are measured from range images obtained by subtraction stereo. Then a method to estimate extrinsic parameters of a camera is proposed. The basic algorithm of the subtraction stereo is implemented on a commercially available stereo camera, and the effectiveness of the method to estimate extrinsic parameters is verified by experiments using the stereo camera.