 |
The goal of this work is to detect obstacles, both positive and negative, through 3D reconstruction, which can be used as a cue for robot navigation. We mounted stereo fisheye cameras on a ground vehicle (robot). Stereo is used to recover depth, and thus to perform scene reconstruction for obstacle detection. Fisheye cameras provide much wider views than regular lens cameras, and the robot can not only "see" what is in front of it, but also "see" what is behind it. This is extremely useful when the robot is in backup mode.
The stereo cameras are calibrated using OCamCalib Toolbox, and the extrinsic parameters are optimized using the Levenberg-Marquardt nonlinear least squares minimization algorithm. We rectify the relevant portion of each omnidirectional image into a virtual perspective image such that epipolar lines are image rows. A semiglobal block matching function (SGBM) is applied to the stereo image pairs to generate disparity images. With disparity, we reproject the image to 3D for scene reconstruction. |
 |
 |
This project describes a framework for segmenting continuous trails in isolated images and tracking them over image sequences for autonomous robot navigation. Proceeding from a shape assumption that the trail region is approximately triangular under perspective, an objective function is formulated in terms of trail appearance, which drives an efficient multi-scale particle filter. A hypothetical trail triangles appearance likelihood is based on a robust measure of color and brightness contrast with and symmetry between flanking triangular regions. The absolute trail likelihood correlates well with confidence that a trail is even visible; the system uses this to switch between appearance cue sets in order to maximize accuracy under changing visual conditions. |
 |
The CPF is used for people tracking in view of the pan, tilt, and zoom (PTZ) camera. The color model of a person is represented by a color histogram. By integrating particle filtering into color-based tracking, the CPF performs better than other color-based trackers when tracking people in cluttered environments, such as partial occlusion and background distraction. |
 |
Spheres have been considered as calibration objects because they have consistent appearance when observed from an arbitrary direction. For this reason, spheres are especially efficient for calibrating multi-camera visual systems compared with other types of calibration objects. The intrinsic parameters of a camera can be recovered based on the relationships between conics of spherical objects and the image of the absolute conic (IAC). A sensitivity analysis of this approach has been performed, and guidelines have been established in order to obtain better calibration results. |