KOSTAS PAPAFITSOROS
Welcome to my personal webpage!
I am a Lecturer in Mathematical Data Science at the School of Mathematical Sciences, Queen Mary University of London.
My research lies in the areas of Mathematical Imaging, Inverse Problems, Variational Regularisation Methods, Optimisation and Deep Learning. I have a second research direction which is the result of my long-term involvement in environmental conservation with special focus on sea turtles, see the corresponding section of this website.
You can contact me at: k (dot) papafitsoros (at) qmul (dot) ac (dot) uk
News
- 12/2025: I was happy to receive a 2025 Impact Research Award from the Faculty of Science and Engineering, Queen Mary University of London.
- 12/2025: New published conference paper Learning spatially adaptive $\ell_1$-norms weights for convolutional synthesis regularisation, with Andreas Kofler, Luca Calatroni and Christoph Kolbitsch published at the Proceedings of the 33rd European Signal Processing Conference (EUSIPCO).
- 10/2025: Together with TurtleWatch 2.0 Egypt (TWE), we have secured an 2025 Innovation Grant from the International Sea Turtle Society. The project will adapt the wildlife photo-ID game "TurtleTurtle" to the sea turtles of the Red Sea using TWE's long-term sea turtle photo-ID data.
- 10/2025: We were successful in securing a £20k EPSRC additional funding for outreach and underpinning the pipeline for AI skills with the project "From pixels to populations: AI in wildlife monitoring". This is together with Eleni Matechou, Linus Wunderlich and Silvia Liverani, all from the School of Mathematical Sciences, QMUL.
- 09/2025: New published paper Nested Bregman iterations for decomposition problems, with Tobias Wolf, Derek Driggs, Elena Resmerita and Carola Schönlieb published at the SIAM Journal on Imaging Sciences.
- 09/2025: New published conference paper WildlifeReID-10k: Wildlife re-identification dataset with 10k individual animals, with Lukáš Adam, Vojtěch Čermák and Lukas Picek, published at the Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).
- 09/2025: New published conference paper Overview of lifeCLEF 2025: Challenges on species presence prediction and identification, and individual animal identification, part of CLEF 2025, Lecture Notes in Computer Science. We contributed via the AnimalCLEF 2025 competition (see here for the working notes).
- 09/2025: I participated at the ENUMATH 2025 conference that took place in Heidelberg, Germany, 1-5 September. Together with Martin Holler and Erion Morina (both from University of Graz), we organised a minisymposium entitled "Learning PDE-based Models from Data" where I also gave a talk.
- 08/2025: We launched "TurtleTurtle" the first interactive educational (and fun!) game about wildlife photo-ID, with this version focusing on the sea turtles of Zakynthos! This was funded by a QMUL Centre for Public Engagement Large Grant. Give it a try!
- 05/2025: We successfully organised the SSVM 2025 conference at Dartington Hall (Devon, UK) on 18-22 May 2025. The conference proceedings (eds: Tatiana Bubba, Silvia Gazzola, Romina Gaburro, Kostas Papafitsoros, Marcelo Pereyra and Carola Schönlieb) can be found here.
- 05/2025: New published paper Exploiting facial side similarities to improve AI-driven sea turtle photo-identification systems, with Lukáš Adam, Cleare Jean, ALan Rees and Vojtěch Čermák. Published at Ecological Informatics.
- 04/2025: I was an invited speaker (online) at iSea's Citizen Science workshop, Athens, Greece.
- 04/2025: New published paper First-order conditions for the optimal control of learning-informed nonsmooth PDEs at Numerical Functional Analysis and Optimization.
- 04/2025: New published conference paper LifeCLEF 2025 Teaser: Challenges on species presence prediction and identification, and individual animal identification, at Advances in Information Retrieval, ECIR 2025, Lecture Notes in Computer Science.
- 03/2025: New preprint Learning spatially adaptive $\ell_1$-norms weights for convolutional synthesis regularisation, with Andreas Kofler, Luca Calatroni and Christoph Kolbitsch.
- 03/2025: Accepted minisymposium: Together with Martin Holler and Erion Morina, we are coorganising a minisymposium on "Learning PDE-based models from data" at the ENUMATH 2025, Heidelberg, Germany, September 2025.
- 03/2025: Together with QMUL's communication team, we just finished publishing a series of 6 outreaching posts under the theme "Sea turtles and Artificial Intelligence". You can read them here: Post 1, Post 2, Post 3, Post 4, Post 5, Post 6
- 03/2025: I was an invited keynote speaker at the annual symposium of the British Chelonia Group.
- 02/2025: Two accepted papers at SSVM 2025 (link to the conference proceedings volumes):
- Deep unrolling for learning optimal spatially varying regularisation parameters for Total Generalised Variation, with Thanh Trung Vu and Andreas Kofler.
- Why do we regularise in every iteration for imaging inverse problems?, with Evangelos Papoutsellis and Zeljko Kereta.
- 02/2025: New accepted book chapter SeaTurtle Photo-Identification: A Practical Guide, with Daphne Hoh, Stephanie Köhnk, Micol Montagna, Josh Witzmann. This is a book chapter due to appear in the upcoming IUCN/SSC Marine Turtle Specialist Group Manual on Research and Management Techniques for the Conservation of Sea Turtles.
- 01/2025: Welcome to Michele Pascale who is starting his PhD with me.
- 12/2024: Together with Lukáš Adam, Lukas Picek and Vojtěch Čermák, we are organising the AnimalCLEF competition focusing on animal re-identification.
- 12/2024: Very excited to receive a QMUL Centre for Public Engagement Large Grant for the project "The wildlife photo-ID game: Can you do better than AI?".
- 11/2024: New published review paper Data-driven methods for quantitative imaging at GAMM-Mitteilungen (with Guozhi Dong, Moritz Flaschel, Michael Hintermüller, Clemens Sirotenko and Karsten Tabelow).
- 11/2024: New published book chapter Machine learning for quantitative magnetic resonance image reconstruction, (with Andreas Kofler and Felix Zimmermann) as part of the book Quantification of Biophysical Parameters in Medical Imaging.
- 09/2024: New preprint Exploiting facial side similarities to improve AI-driven sea turtle photo-identification systems (with Lukas Adam, Claire Jean and ALan Rees).
- 09/2024: New preprint Nested Bregman iterations for decomposition problems (with Tobias Wolf, Derek Driggs, Elena Resmerita and Carola Schönlieb).
- 06/2024: New preprint WildlifeReID-10k: Wildlife re-identification dataset with 10k individual animals (with Lukáš Adam, Vojtěch Čermák and Lukas Picek).
- 06/2024: I gave an online talk at the French-German-Spanish Conference on Optimization 2024 at the minisymposium "Optimization methods for inverse problems and beyond".
- 05/2024: I was at the SIAM Conference on Imaging Sciences, Atlanta, US, where I coorganised a minisymposium on Deep Unrolled Optimisation Methods for Inverse Imaging Problems with Evangelos Papoutsellis and Andreas Kofler.
- 05/2024: I gave a talk at the workshop on Big Data Inverse Problems, ICMS Edinburgh.
- 05/2024: Welcome to Thanos Polychronou who is starting his PhD with me.
- 04/2024: Excited to receive two grants to support and enhance the Zakynthos Turtles project during the 2024 season: A QMUL Centre for Public Engagement Small Grant as well as a Conservation Grant from the British Chelonia Group.
- 04/2024: New preprint Data-driven Methods of Quantitative Imaging (with Guozhi Dong, Moritz Flaschel, Michael Hintermüller, Clemens Sirotenko and Karsten Tabelow).
- 03/2024: New accepted paper A descent algorithm for the optimal control of ReLU neural network informed PDEs based on approximate directional derivatives (with Guozhi Dong and Michael Hintermüller) at the SIAM Journal on Optimization.
- 03/2024: New preprint Machine Learning for Quantitative MR Image Reconstruction (with Andreas Kofler and Felix Zimmermann).
- 01/2024: Two published papers at the Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (both with Lukáš Adam, Vojtěch Čermák and Lukas Picek):
- WildlifeDatasets: An Open-Source Toolkit for Animal Re-Identification (best paper award!)
- SeaTurtleID2022: A Long-Span Dataset for Reliable Sea Turtle Re-Identification.
