Tom Drummond
Magic table
The aim of this project is to create a table which can move objects placed on its surface in arbitrary directions. This will be achieved by shaking a thin sheet that sits on the table (a tablecloth) using voice coils. By shaking the cloth in the right way it is possible to move an arbitrary number of objects in arbitrary directions.
ARM optimised computer vision library
The aim of this project is to create a library of image processing and computer vision algorithms that makes use of the ARM Neon SIMD unit and renderscript to create high performance components.
Android robotics
Not humanoid - but using smart phones that run the Android operating system. The aim of this work is to develop one or more robotic platforms which use a smart phone such as the HTC Desire as the robot brain connected to motor and servo controllers. The phone will also capture video, process it and use this to control the robot.
Kinect-based tracking
The aim of this project is to develop a visual tracking system that acquires its 3D model of the environment in real time from a Microsoft Kinect sensor. This will be used to identify keypoints and their locations in 3D for matching to a conventional camera view that moves through the environment.
Hand gesture recognition
The aim of this project is to identify hand pose and gestures from a pair of video cameras using templates of appearance. For pointing gestures, the direction of the pointing finger must also be established.
Surface computing
The aim of this project is to establish a surface computing environment which uses cameras to observe a tabletop and a data projector to add information to objects on the table’s surface. This is important for augmenting paper-based media such as maps and for tangible user interfaces.
The aim of this project is to create a table which can move objects placed on its surface in arbitrary directions. This will be achieved by shaking a thin sheet that sits on the table (a tablecloth) using voice coils. By shaking the cloth in the right way it is possible to move an arbitrary number of objects in arbitrary directions.
ARM optimised computer vision library
The aim of this project is to create a library of image processing and computer vision algorithms that makes use of the ARM Neon SIMD unit and renderscript to create high performance components.
Android robotics
Not humanoid - but using smart phones that run the Android operating system. The aim of this work is to develop one or more robotic platforms which use a smart phone such as the HTC Desire as the robot brain connected to motor and servo controllers. The phone will also capture video, process it and use this to control the robot.
Kinect-based tracking
The aim of this project is to develop a visual tracking system that acquires its 3D model of the environment in real time from a Microsoft Kinect sensor. This will be used to identify keypoints and their locations in 3D for matching to a conventional camera view that moves through the environment.
Hand gesture recognition
The aim of this project is to identify hand pose and gestures from a pair of video cameras using templates of appearance. For pointing gestures, the direction of the pointing finger must also be established.
Surface computing
The aim of this project is to establish a surface computing environment which uses cameras to observe a tabletop and a data projector to add information to objects on the table’s surface. This is important for augmenting paper-based media such as maps and for tangible user interfaces.
