I am a zoologist / experimental psychologist whose research interests primarily concern computational approaches to applied vision and questions lying at the intersection of sensory biology, psychology, history and art. I am particularly passionate about how visual scenes can be “understood” using computer vision and what comparisons can be drawn with biological visual systems.
Projects I work on:
Automatic disease detection and monitoring in calves
I am an EPSRC Innovation Fellow on a project developing automatic disease detection and monitoring in domestic cattle calves. The project uses artificial intelligence techniques, coupled with visible-range and thermal cameras, to identify Bovine Respiratory Disease (BRD), one of the most common and costly diseases affecting cattle in the world, at the earliest possible stage. My work includes building sensors, deploying and maintaining equipment on farms, establishing efficient data transfer protocols and analysing data using deep neural networks.
An ongoing research to develop a toolkit and pipeline in order to establish the best (or worst) camouflage for any object in any environment for any viewer using deep neural networks. A preprint on establishing optimal concealing and warning colouration can be found here.
Modelling the evolution of camouflage using Generative Adversarial Networks
This research, supported by Nvidia, investigates how competing deep neural networks can be used to evolve camouflage patterns. Our team has demonstrated how Generative Adversarial Networks (GAN) can be utilised to simulate an evolutionary arms-race between a synthetic prey and predator.
Understanding art reception from eye movements using AI-generated artworks
The project examines how we can predict preference to visual artworks using eye movements of observers and artworks generated by deep neural networks.
Cultural evolution of military camouflage uniform patterns
My PhD work focused on how camouflage uniform patterns evolved since the early 20th century. The research uses methods from computer vision to establish similarity metrics between patterns and phylogenetics to model how patterns of allied / hostile countries have influenced each other’s designs.
I graduated from the University of Bristol in 2011 with a BSc (Hons) in Psychology & Zoology. I did my final year dissertation on gloss perception in the large earth bumblebee (Bombus terrestris). I have decided to stay in Bristol to do a PhD; my thesis investigated the cultural evolution of camouflage uniform patterns and was supervised by Prof. Innes Cuthill (School of Biological Sciences), Prof. David Bull (Department of Electrical and Electronic Engineering) and Dr Gavin Thomas (Department of Animal and Plant Sciences, University of Sheffield). I have been previously working as a Research Associate on the EPSRC-funded "Camouflage Machine" project, which combined machine learning, visual psychophysics and computer graphics to either optimise camouflage or maximise visibility for a range environments.
Talas, L., Fennell, J. G., Kjernsmo, K., Cuthill, I. C., Scott-Samuel, N. E., Baddeley, R. J. Evolving optimum camouflage with Generative Adversarial Networks. bioRxiv, 429092.
Fennell, J. G., Talas, L., Baddeley, R. J., Cuthill, I. C., Scott-Samuel, N. E. Optimising colour for camouflage and visibility: the effects of the environment and the observer’s visual system. bioRxiv, 428193.
Cuthill, I. C., Allen, W. L., Arbuckle, K., Caspers, B., Chaplin, G., Hauber, M. E., Hill, G. E., Jablonski, N. G., Jiggins, C. D., Kelber, A., Mappes, J., Marshall, J., Merrill, R., Osorio, D., Prum, R., Roberts, N., Roulin, A., Rowland, H., Sherratt, T. N., Skelhorn, J. Speed, M. P., Stevens, M., Stoddard, M. C., Stuart-Fox, D., Talas, L., Tibbetts, E. & Caro, T. 2017. The biology of color. Science, 357, eaan0221.
Talas, L., Baddeley, R. J. & Cuthill, I.C. 2017. Cultural evolution of military camouflage. Philosophical Transactions of the Royal Society B: Biological Sciences, 372(1724), 20160351.
Talas, L. & Talas, L. 2017. Infrared thermography as an imaging diagnostics tool for equine medicine. Hungarian Veterinary Journal, 139, 259-268.