Research
Current projects
Unlocking the complexity of organismal colour patterns using artificial intelligence
Partners: Cooney (PI, Biosciences), Thomas (Biosciences), Maddock (Computer Science), Han (Computer Science). This project is funded through the 23IPAP BBSRC 2023 International Partnering Awards Plus on AI for Bioscience. The aim is to develop a powerful and generalisable toolkit for extracting and analysing colour pattern information from biological images and thus transform the ability of researchers to rapidly characterise colour pattern phenotypes from image datasets.
Neural Style Transfer
Most work on neural style transfer concentrates on images, with applications such as Prisma for turning photos into art. This PhD project, which began in October 2020, is focussing on the use of depth in neural style transfer, which can be applied to images, videos and computer games. More information: https://ioannoue.github.io/
Wet cloth
There are multiple simulation approaches for both liquids and cloth. Combinations are less common. This work is considering multiphase combinations. More information: https://robden820.github.io/
Text-guided 3D object production
There has been a lot of recent attention on using text input to produce images. This project is investigating how to use text input to produce 3D objects. More information: https://hm1st.github.io/