My research specialises in the fields of Computer Vision, Machine Learning, Variational Methods and Partial Differential Equations. The focus of my current research is the development of novel methods for detailed 3D modelling and reconstruction of real-world objects from image data. The main motivation for my work stems from the vision of building robust 3D Computer Vision methodologies that would provide highly-accurate estimations in challenging real-world scenarios.
In more detail, my research interests include dense variational methods, convex and nonconvex optimization, statistical shape models and deep learning for 3D face modelling, dense 4D (3D+time) reconstruction of non-rigid scenes, motion estimation and more general inverse problems in Computer Vision.
I am interested in various applications such as gesture and sign language recognition, computer-aided surgery, marker-less motion and performance capture, augmented reality, visual effects, facial expression recognition, and biomedical image processing.
For more details, please visit the section Research Topics of this website.
Also, this is a (rather outdated) demo video with an overview of my research up to March 2017:
Ongoing Research Grants
FSM (2022-2024): The Forth Skull Model project Funding source: Internal Seed Funding Grants, Institute of Computer Science (ICS), FORTH. Role: Principal Investigator (PI).
STEAMDIVE (2022-2025): Diversity in STEAM Funding source: Erasmus+, European Union. Role: Member of FORTH's research team, leading the research on interactive installations for real-time 3D face reenactment.
NVIDIA Academic Hardware Grant (awarded in 2022): 3D-aware Neural Rendering for Photo-realistic Synthesis of Sign Language Videos. Funding source: NVIDIA. Donation of high-end GPU hardware (RTX A6000 48GB) to support the relevant resarch. Role: Principal Investigator (PI).
Past Research Grants
SignGuide (2020-2023): Automatic guidance in museums by using sign language Funding source: Hellenic Foundation for Research and Innovation (HFRI), Greece. Role: Member of FORTH's research team, conducting research on modelling and recognition of non-manual signals.
FACER2VM (2016-2020): Face Matching for Automatic Identity Retrieval, Recognition, Verification and Management Funding source: Engineering and Physical Sciences Research Council (EPSRC), UK (Programme grant). Role: Member of Imperial College's research team, leading the research on dense 3D modelling and reconstruction of facial shape from monocular input.
GOSH 3D Morphable Models grant (2014-2016) Funding source: Great Ormond Street Hospital (GOSH), UK. Role: Member of Imperial College's research team, leading the research on applying large-scale 3D face modelling to corrective surgery on patients with facial deformities.
HUMANIS (2008-2014): Human Motion Analysis from Image Sequences Funding source: European Research Council: ERC Starting Grants. Role: Member of Prof Agapito's research team, leading the research on dense dynamic 3D reconstruction of generic non-rigid scenes from a single camera.