Here are my project suggestions for PartII or Diploma students in the academic year 2003/2004. Some of the information on last year's suggestions may also be relevant. I may also consider supervising projects in other areas that I am familiar with such as video tracking, object recognition, machine learning and inference.
The platform of choice for implementation of the projects is Matlab, which is available in most Colleges and in the CL. Matlab has excellent facilities for numerical computations and visualisation, and there are many useful toolboxes (e.g. for image processing, statistics, optimisation, neural networks). For reasons of runtime efficiency it might however be appropriate to implement part of the required functionality in a lower level compiled language such as C++ and integrate such modules into Matlab by means of the Matlab compiler package. There are various free computer vision packages available which use C/C++. Probably the best (in terms of features and support) of these are Intel's OpenCV and the VXL libraries.
No previous experience of image and video processing is required, just enthusiasm. The projects are challenging in that they address interesting research problems, but plenty of support will be available. Apart from an interest in the project, a reasonable grounding in continuous mathematics and probability theory would be helpful, as would proficiency with high level programming languages such as C++ or the Matlab environment.
As regards general references, I particularly recommend the textbook by Forsyth and Ponce on computer vision. References on image processing, such as the book by Gonzales and Woods, and on numerical methods, such as Numerical Recipes in C++, might also be handy. Useful online resources for computer vision include CVonline and the Computer Vision Homepage at CMU. One of the best tools for finding papers etc. apart from Google is Citeseer. There are many online tutorials for Matlab, this is a local one at CUED. Eventually you might want to use Tex/LaTex to produce your dissertation, here is a basic introduction and further information can be found here and here.
Methods for tracking objects from image sequences often rely on some model of appearance in order to consistently identify objects of interest without manual intervention. This project will implement and analyse methods for modelling appearance and appearance changes for object tracking in a way which is robust to the presence of noise, occlusions, and changes in position, motion, orientation, lighting, and background. Traditionally many approaches in computer vision have suffered from a trade-off whereby increasing robustness leads to a loss of generality of the proposed methodology. The aim of the project is to show how recent advances in statistics and machine learning can be harnessed to provide frameworks for appearance modelling which exhibit the desired properties of robustness while remaining applicable across a broad range of tracking contexts provided that there is enough training data to characterise the domain of interest.
Please feel free to contact me if you are interested in this project proposal and would like to know more. For additional information, have a look at the following:
There is much evidence in artificial and biological vision which supports the assertion that edges and other discontinuities in the visual spectrum are a vital source of information for a wide variety of perceptual tasks such as figure-ground segmentation, motion and shape analysis, and depth perception. This project will investigate and implement some recently proposed methods which utilise edge based information (e.g. parameterised contours or sample points) in order to facilitate the visual tracking of features, regions, and entire objects.
Please feel free to contact me if you are interested in this project proposal and would like to know more. For additional information, have a look at the following:
Recent research has shown that fairly simple geometric feature classifiers arranged in cascades and trained using machine learning methods such as boosting outperform other component or appearance based approaches to face detection, both in terms of accuracy (false positive and false negative rates) and speed. This project will implement a framework such as that proposed by Viola and Jones (see Rapid Object Detection using a Boosted Cascade of Simple Features and Empirical Analysis of Detection Cascades of Boosted Classifiers for Rapid Object Detection). Emphasis will be placed on correct implementation and performance analysis of such an algorithm (or a variation thereof) rather than invention of a completely new method.
Please feel free to contact me if you are interested in this project proposal and would like to know more. For additional information, have a look at the following:
I am collaborating with the sentient computing project at the Laboratory for Communications Engineering (LCE). This system uses mobile ultrasonic sensor devices known as bats and a receiver infrastructure to gather high-resolution location information for tagged objects such as people and machines in order to maintain a sophisticated software model of an office environment. Applications can register with the system to receive notifications of relevant events to provide them with an awareness of the spatial context in which users interact with the system.
My own research is focused on utilising the sentient computing infrastructure for multi-modal fusion (i.e. of video analysis results and ultrasonic location events) and for the creation of annotated training data. Possible applications of this work which would be suitable as projects include vision-based biometrics (e.g. face and gait recognition) for verifying user's identity, vision aided intruder detection for sentient environments, and using vision to augment the sentient computing system's capabilities for HCI.
In addition, there are other aspects of the sentient system which would be suitable for partII or Diploma projects. Possible applications include a system for providing navigational assistance in large buildings (such as the University Library), novel personalised human computer interfaces, and robotic control on the basis of ultrasonic localisation and tracking.
If you are interested, please look at the sentient computing webpages and contact myself or Rob Harle at the LCE.