Sponsored by The Foundry.
My project aims to develop efficient techniques for automatically editing and analysing digital film. The main areas of interest are colour grading, disparity mapping and focus detection. In colour grading, the colours, brightness and contrast properties of a sequence are altered both for technical reasons, for example to match sequences shot at different times and in different lighting conditions, and in order to achieve a visual style and mood. Disparity maps are central to stereo vision, and in film production allow for combining 3D graphics with stereo footage.
Sponsored by Network Rail
The Thameslink Project involves a £5 billion upgrade of a major railway line through Central London. The protection of existing buildings and activities as well as major transport infrastructure hubs (e.g. Kings Cross/St Pancras, Blackfriars, Farringdon and London Bridge) is essential to making such developments sustainable as transport solutions. Monitoring is a fundamental requirement for safety, protection and operational efficiency. Current railway monitoring activities involve placing targets by drilling, clamping or gluing directly onto the structures which can be intrusive, create safety issues as well as be time consuming and expensive.
Advances in survey techniques, particularly with laser scanning and photogrammetry, offer the possibility to remotely capture vast amounts of data quickly without the need of direct surface contact. This could significantly reduce the costs and time expenses, as well as safety issues currently involved. In parallel new visualisation and mixed reality computing technologies offer the potential for rapid on-site confirmation of changes rather than the numeric feedback of current techniques.
My project will look into key issues for the application of new spatial measurement techniques including the accuracy and precision achievable, handling the data volumes generated and the ability to rapidly analyse and communicate results. This will be done by investigating a selection of current railway monitoring situations and to segment their existing instrumentation with new technologies. Careful design using numerical simulation in combination with real data will allow quantitative and qualitative comparisons of the capabilities of different techniques and lead to the development of best practice for their adoption (including cost/benefit analysis).
Sponsored by LottoLab
In order to determine the effect of ecology on the emergence of brain and behaviour, it is necessary but not sufficient to examine the rich and varied strategies adopted by the animal kingdom. But we must also look beyond the animals and model the ecology that formed them. This project will develop a novel approach to modelling that focuses on the ecology of visual agents and their adaptation to their ecology. This approach will proceed by building model ecologies that can be systematically varied, in order to see which features lead to the emergence of some of the stunning features of visual perception. The project draws on mathematical and computational modeling, and may include psychophysical or behavioural experiments in order to validate the models.
Sponsored by Halcrow
An understanding of interconnected and complex hydrological systems is necessary for sustainable water management and engineering solutions (e.g. waste water treatment facilities and flood defences), especially in light of global environmental change. Increasingly, computer models of individual natural and artificial features such as precipitation, runoff, rivers, groundwater, waste water and estuaries are being coupled to enable a more holistic understanding of how water moves through entire river basins – from cloud to coast. In parallel, there is a need to communicate these complex interrelationships at a conceptual level to “lay” audiences including other engineers, policymakers, and the public.
To date, however, hydrological visualisations have typically focused on displaying the behaviour of individual hydrological features or subsystems. The C2C visualisation project brings these subsystems together to create holistic 3D river basin visualisations, thereby allowing non-experts to understand interconnected natural and artificial water systems at a conceptual level. The visualisations will remain “true to life” through validation against real-world and model simulation datasets.
John HindmarchSponsored by the Science Museum.
My project is concerned with both the technical aspects of scanning museum objects, and the use of those scans in a physical or virtual exhibition space. I am particularly interested in the practicalities of providing an interface through which the general public can interact with 3D models, and the attitudes of the public towards exhibitions of virtual objects.
Kazim PalSponsored by London Metropolitan Archives
This project deals with digitally restoring parchments with fire damage. The particular parchment book is an extremely important document that fell victim to this very form of damage. This project is planning to use computer vision and computer graphics techniques to develop a digital acquisition and reconstruction procedure that would help to make the parchment book accessible again. A particular goal is to devise an acquisition apparatus and method to digitally flatten the individual pages, exposing their content in a legible form.
Support by the UCL Overseas Research Scholarship
Whilst current video conferencing systems can support collaboration at a distance, even high-end systems using high-definition video fail when the users don't pay attention to the spatial configuration of the screen, cameras and the users' own positions. The most obvious way that this manifests is when users misunderstand each others' eye gaze. A new class of video conferencing system is emerging that use multiple video views, and which can track user's movements. In this research we will investigate how a multi-view, multi-site video conferencing system can exploit eye-tracking information. The main technical work will be in building example teleconferencing systems using high-definition video which incorporate eye-tracking. There will also be substantial human-computer interaction-based analysis of group interaction.
Sponsored by Disney Interactive
The aim of my project is to give game design artists better 3D references and scene reconstructions which can be directly fed into the creative pipeline. This is motivated by the increasing public demand for detailed, complex 3D worlds and the resulting demand this places on game design artists.
Sponsored by PLP Architecture
PLP Research is focused on providing better tools for architects and engineers to predict behaviour of complex systems, such as the built environment. The control of geometric form has been the focus of current parametric design systems. By teaming up with UCL Energy Institute and Bentley Systems, PLP aims to extend that ease of control into the energy space-time domain, where manipulating performance is as important as manipulating form. By coupling geometric form and performance very tightly, optimisation and exploration will be orders of magnitude faster and easier. The purpose of this collaboration is to develop and test a parametric design environment which addresses the issues raised by environmental and sustainable design. This collaboration aims to extend an established parametric environment, Generative Components by Bentley Systems.
Sponsored by Feilden Clegg Bradley
I am researching the cases for refurbishment and replacement of existing buildings to improve whole life cycle energy consumption and reduce carbon emissions.
The energy performance of an old building is typically poorer than that of newer buildings. Such buildings make up a large proportion of the existing building stock and if they remain in use without improvement will form an increasingly significant and irreducible proportion of the overall energy consumption. There is a compelling argument for replacing existing buildings with newer, more energy efficient buildings. However this should be measured against the significant energy invested in the construction of a replacement building, particularly in terms of the embodied energy of the materials used and the construction process. This is in addition to the many other implications of replacing a building.
I plan to apply Life Cycle Assessment methodology to a variety of case studies to explore the impact of refurbishment or replacement on life cycle energy and carbon emissions. My target outcome is a software tool to be used to aid decision making on refurbishment versus replacement of a building.
Page 1 of 2«StartPrev12NextEnd»
Sponsored by Laing O'Rourke
Emerging technologies in digital fabrication are rapidly changing the way we can build, and opening up new possibilities for a performative and sustainable architecture. Freeform Construction develops technologies for digital manufacturing and design methods geared towards an architectural 3D printing, or additive manufacturing, solution.
My work is focused on developing design methods and solutions which enable the potential of an additive manufacturing paradigm -mass customisation and a vast formal complexity and variation. In these scenarios algorithmic design methods can allow the structure of a building to start responding to their context in previously unattainable ways and what has been considered separate scales in space and time can be fully integrated. By learning from natural processes we are striving to create structures which are functional membranes and which allow us to bypass the dependency on external energy to control heat, humidity and ventilation. The membrane utilises local energy gradients in the surrounding environment - for example light, turbulent winds - to drive and control the exchangeRead more
Dave di Ducca
Sponsored by Jason Bruges Studio
When an illusionist or magician is performing a trick, they modify their performance based on the actions of the observer. Techniques such as ‘multiple outs’ enable the illusionist to tailor the situation they are designing, in reaction to the observer – in real time. The interactivity of the situation facilitates the imagination of the observer to influence the designing of the situation itself.
My research suggests that, as experience is a perceptual entity, experience cannot be designed. However, like magicians, designers are able to explore the individuality of observers’ spatial experiences through inferring both observer attributes and changes in attributes from simple observations. Contrary to most static objects and buildings, this is a situation which can modify the affordances it presents in response to real time observations of the character of the observer. I aim to develop spatial situations with similar qualities.
Concept designs are key to the direction and ultimate success of a project. Major decisions which fundamentally affect the sustainability of a building are made in the early stages of design. However, due to the rapid and erratic development of design, environmental analysis is not usually undertaken until after the concept stage. By this time, most sustainable solutions would need design or briefing changes that are often too costly to implement.
Aedas R&D develop design methods through modelling and computation that quantify, visualise and manipulate spatial, environmental and financial factors of design. In collaboration with Aedas R&D and UCL, my research will explore the use of advanced modelling at the early stages of design, utilising performance data and environmental simulation as architectural design drivers. I aim to explore alternative spatial designs and form-finding techniques while developing the methodology for integrating quantitative analysis into the design workflow that present the performance implications of early design decisions parametrically. The objective is to develop a system that provides quick feedback while designers are undertaking initial exploratory exercises. Read more
The main aim of this project is to quantitatively determine the operational performance of newly built, low-energy schools. This will be achieved by examining discrepancies between the design and the actual building in four predominant areas; building energy performance, building systems installation, building use, and the internal environment performance. While each of these areas can be investigated separately, in order to provide a complete understanding of the building the interrelationship of these areas has to be discerned.
Through utilising current post-occupancy evaluation tools and developing new procedures and methods for evaluating the performance of schools, it is hoped that the reasons for the discrepancies will become clear. It is this development of new procedures and methods of post-occupancy evaluation that will form the core of this project. These new tools will aim to not only review the quantitative aspects of building design, but also the more qualitative aspects as well, such as user satisfaction and quality of the learning environment.Read more
Sponsored by Foster + Partners
My research concerns the visual experience of architecture and urban space. I will be exploring the use of machine perception to understand how visual factors influence spatial awareness. Using this research as a foundation, I intend to further investigate how such visual qualities can be simulated in the generation and analysis of architectural design options.