Information

Name : Gwyneth Bradbury

Email : g.bradbury@cs.ucl.ac.uk

Location : Room 4.17, Malet Place Engineering Building, UCL

---

Research Information:

I am an EngD VEIV student at UCL having already completed the Msc Computer Graphics, Vision and Imaging here. My project looks at creating a tool for virtual environment design artists to increase work flow.

Acquiring Procedural Environmental Models in The Wild

The main aim of the project is to give games design artists better 3D references and scene reconstructions which can be directly fed into the creative pipeline. This is motivated by the increasing oublic demand for detailed, complex 3D worlds and the resulting demand this places on game design artists.

This project is looking at developing acquisition and modelling technologies that provide more than just a visual reference: in the context of this project, visual acquisition and reconstruction methods shall be developed that provide richer, three-dimensional references, and that ultimately yield scene reconstructions that can directly be fed into the content creation pipeline. The project will focus on natural environments (as opposed to urban scenes) and may combine multi-spectral imaging, wide-baseline stereo reconstruction and semantic scene analysis to obtain approximate procedural representations of natural scenes.

Current Projects

The first year of the EngD VEIV program involves several projects, aiming both for inter-disciplinary collaboration as well as a solid start to personal research. My current projects are:

• 3D reconstruction for natural scenes:

  This involves looking at different means of scene reconstruction with particular application to natural environments and is as much a programming exercise as a research initiative.

• Group Project, Machine-Learning for Updating Parametric Computational Fluid Dynamics Models:
  This project will apply Machine Learning Techniques to computational fluid dynamics models with the aim of creating a system capable of predicting turbulence around a structure and optimising the structure to reduce this turbulence.

I am also taking courses in Supervised Learning and Graphical Models.

Past Work

As part of my MSc here, I undertook a project on semantic scene Analysis of natural scenes using 2D images. The results weren't entirely discouraging but vast improvements in the Machine Vision techniques that were and could be used are needed to improve the results and extend them to 3D images/reconstructions. Some Past Work can be seen here (thesis available on request).

Computer Science Student Conference 2011

EngD VEIV Symposium 2012

Image-Based Rendering Lecture videos