A concept map is essentially a graph where edges indicated the relations between the nodes (sometimes also a hyper-graph having edges with more than two nodes). The idea of this project is to use concept maps to build an advanced desktop environment. A node in the concept map corresponds to a file and files can be grouped via a single hyper-edge into belonging to a sub-set of the graph as a generalisation of directories. In this scheme a file can belong to more than one of these subsets. The user can add additional nodes to represent other relationships between files (either only between nodes belonging to the same subset or adding new edges may create new subsets). Based on the subset relations a concept of collapsing or expanding these subsets in the view can be developed to create a hierarchical view rather than always viewing the complete graph. Part of this project is to explore how to best utilise concept maps for a desktop. A suitable representation for the concept map has to be developed, basic operations on the concept map (subset grouping, adding edges and nodes, deleting edges/nodes, creating subsets, etc.) should be provided and a 2D or 3D rendering engine for these graphs is required.

The aim of the project is to understand how 3D graphics are created in computers and basics of artificial intelligence. The type of game that will be created will be an RPG. The scope of the project will need to be limited as a full game will not be practical in the time period for the project.

I want to produce a driving simulation of a manual gearbox car using the Xbox 360 joypad, with the trigger buttons as the pedals; the L and R triggers being the clutch and accelerator respectively, the brake will be the Right Bumper button. The Left Stick will change gears and the Right Stick will be used to steer. The game will probably by a time trial style, though online multiplayer is supported by XNA and so will be researched.

Taking multiple photos of a single scene with different settings and modalities allows to merge these photos in a post processing step into a single image, either with improved quality or other special effects. This project is aimed at exploring some of these techniques. In particular: (i) high dynamic range imaging allows a greater dynamic range of luminance between light and dark areas of a scene than normal digital imaging techniques (represent the wide range of intensity levels found in real scenes ranging from direct sunlight to shadows by combining photos taken with different exposure settings and adjusting local contrast). (ii) some bilateral filtering variants combine multiple photos to remove noise, in particular from low-lighting scenes (improve the low-light image with information from, e.g., a flash photo). Optionally combining photos with 3D range scanner information to process the image could also be explored. Requires programming skills in Java or C/C++ or python or matlab and good image processing skills.

A java application of a turn based strategy game with some similarities to a chess style game. The game is played on a grid and two players take it in turn to move their pieces. Pieces are selected before the game (and possibly at certain points during) from a selection. This is based on points where better pieces cost a greater number of points than less powerful pieces. Difference pieces will have different abilities such as the way they move and how they take the opposition pieces. It will be played with two players taking turns or one player verses the computer AI. Also with the option to include network play for two players over a network or the internet