Head Tracking Driven Virtual Computer Mouse


Introduction:

Recent new applications in Human-Computer Interaction (HCI) and Computer Vision (CV) bring a great opportunity to the improvement of human life. The future desired Perceptual User Interfaces (PUI) and human inter-face devices require more reliable and fast system performance in dealing with versatile use cases. Much attention has been focused on the scenario of tracking user’s movements with a video camera and translating the motion parameters into semantic indicator symbols to manipulate machine operation. Camera based perceptual user system provides an alternative solution for convenient device control, which encourages the application of very low bit-rate video-teleconferencing, interactive computer games, machine guided therapeutic gymnastics, robot control, etc. It also significantly helps the disabled and elder people use limited voluntary actions to communicate with others .
Camera mouse system is such a representative hand-free PUI that tracks human body movement in video to manipulate virtual computer pointing devices to respond user intentions. In particular, since human head and face embody the focus of attention, head/face tracking based hand-free computer mouse is a more intuitive and convenient PUI solution. Great progresses in the field of face detection/ tracking and recent advances in hardware have made it possible of building real-time tracking system on common PCs.
Our idea is to use a robust and reliable real-time head tracker for tracking, estimate the course head pose and relative motion simultaneously, and translate the motion parameters to control the virtual computer mouse. To overcome the drawback of single module system, we integrate the rapid face detection and image-based head tracking as well as design a transition strategy for mode switch when one module fails.


Existing System:
We now use optical mouse and touch pads for navigating through personal computers and laptops. But if the optical mouse or the touch pad that we use experience a technical break down then there is no way we could do the navigation on the system. Key board may prove to be

inefficient in this case as we are highly dependent on these tools. We could operate a system whose keyboard is malfunctioning with a mouse but the other way around is really not easy. It is also not easy for elders and disabled person to use the mouse. Our proposed system overcomes this difficulty by making use of Human-Computer Interaction (HCI) and Computer Vision (CV) which has proven to be a milestone in the field of science and technology.

Proposed System:
A novel head tracking driven camera mouse system, called "hMouse", is developed for manipulating hand-free perceptual user interfaces. The system consists of a robust real-time head tracker, a head pose/motion estimator, and a virtual mouse control module. For the hMouse tracker, we propose a 2D detection/tracking complementary switching strategy with an interactive loop. Based on the reliable tracking results, hMouse calculates the user's head roll, tilt, yaw, scaling, horizontal, and vertical motion for further mouse control. Cursor position is navigated and fine tuned by calculating the relative position of tracking window in image space and the user's head tilt or yaw rotation. After mouse cursor is navigated to the desired location, head roll rotation triggers virtual mouse button clicks. Experimental results demonstrate that hMouse succeeds under the circumstances of user jumping, extreme movement, large degree rotation, turning around, hand/object occlusion, part face out of camera shooting region, and multi-user occlusion. It provides alternative solutions for convenient device control, which encourages the application of interactive computer games, machine guidance, robot control, and machine access for disabilities and elders.

The main contributions and advantages of our work are summarized as follows.
1.We propose a new camera mouse system.
2.A robust and fast tracker is developed.
3.A low- cost head pose and motion estimation scheme is introduced.
4.We design an efficient strategy for virtual mouse manipulation in real-time. Our hMouse has all basic functions of hand-hold computer mouse, which can control Microsoft WINDOWS OS and play games.

Implemented on a typical PC our hMouse provides a specific virtual human interface for
hand-free mouse connection. Like the general camera mouse, hMouse is composed of a virtual tracking module and a mouse control module. By using a common consumer camera without calibrated lens, the system processes each frame of the captured video in real time. User’s face/head is first automatically detected and tracked by a robust and reliable head tracker. The head pose and motion parameters are further estimated by analyzing visual cues. With basic synchronization control and temporal smoothing, visual tracking module navigates cursor and control virtual mouse buttons using the received motion parameters. Operating system finally responses all mouse events generated by this PUI.

Software Specification :
Front end : J2EE
Back end : My Sql
Operating System : Windows/ Linux
IDE : Net Beans
Hardware Specification :
Processor : Pentium IV

System Bus : 32 BIT

RAM : 512 MB

HDD : 40 GB

Display : SVGA Color
Key Board : Windows/Linux Compatible
[attachment=4181]