In the soccer broadcasting, human camera operators adjust the pan angle of a pan, tilt and zoom (PTZ) camera to follow player positions and ball position. In this project, we assume the camera only has 1 degree of freedom (DoF) which is the pan angle. And the player positions are the main targets when the cameraman shots the video. The project aims to analyze the relationship between the pan an...

The appearance of individuals captured by multiple non-overlapping cameras varies greatly due to pose and illumination changes between camera views. In this paper we address the problem of dealing with illumination changes in order to recover matching of individuals appearing at different camera sites. This task is challenging as accurately mapping colour changes between views requires an exhau...

The current visual simultaneous localization and mapping (SLAM) systems require the use of matched feature point pairs to estimate camera pose construct environmental maps. Therefore, they suffer from poor performance matchers. To address this problem, a SLAM using deep matcher is proposed, which mainly composed three parallel threads: Visual Odometry, Backend Optimizer LoopClosing. In extracto...

Nonlinear diffusion filtering and wavelet/frame shrinkage are two popular methods for signal and image denoising. The relationship between these two methods has been studied recently. In this paper we investigate the correspondence between frame shrinkage and nonlinear diffusion. We show that the frame shrinkage of Ron-Shen’s continuous-linear-spline-based tight frame is associated with a fourt...

State of the art visual SLAM systems have recently been presented which are capable of accurate, large-scale and real-time performance, but most of these require stereo vision. Important application areas in robotics and beyond open up if similar performance can be demonstrated using monocular vision, since a single camera will always be cheaper, more compact and easier to calibrate than a mult...

We present a novel solution for extrinsically calibrating a camera and a Laser Rangefinder (LRF) by computing the transformation between the camera frame and the LRF frame. Our method is applicable for LRFs which measure only a single plane. It does not rely on observing the laser plane in the camera image. Instead, we show that point-to-plane constraints from a single observation of a V-shaped...

This paper describes a new method for detection and tracking of moving objects from a moving camera image sequence using robust estimation and active contour models. We assume that the apparent background motion between two consecutive image frames can be approximated by affine transformation. In order to register the static background, we estimate affine transformation parameters using LMedS (...

This paper presents results demonstrating real-time six degree of freedom localization, mapping, navigation and obstacle avoidance in an outdoor environment using only a lowcost off-the-shelf inertial measurement unit and a monocular camera. This navigation system is intended for operation of small unmanned aerial vehicles when GPS signals are unavailable due to obstructions or jamming and when...

Dual systems of steel moment frame and reinforced concrete shear wall have combined the advantages of steel frames and reinforced concrete shear wall. These walls have increased the lateral stiffness of steel frames and have reduced seismic demands on steel frames thus providing opportunities to use such system. In this research intermediate dual system of steel moment frame was chosen with int...

We propose a robust algorithm for tracking camera motion in unconstrained urban environments. We achieve this by using both an attached laser scanner, which gives sparse 3D structure for each frame, and image-based correspondences from a video camera. The combination of a set of 3D points and their associated projections in each frame can be used to find the motion between frames. We propose a ...