Melinos Averkiou

Group leader @ CYENS Centre of Excellence ‚ÄĘ Senior Research Scientist @ University of Cyprus

prof_pic.jpg

Welcome to my personal website !

I am currently the group leader of the Visual Computing Group (VCG) at CYENS Center of Excellence, where I lead a team of research scientists working on deep neural networks for solving tasks in scene understanding and visual recognition. I am also a Senior Research Scientist in the Department of Computer Science of the University of Cyprus, where I carry out research, supervise students and teach courses in the broad area of visual computing.

My research spans the areas of computer vision, machine learning, and computer graphics, focusing on data-driven methods for learning semantics from 3D and 2D data. I am particularly interested in deep neural network architectures for acquiring, modelling, and understanding the semantics of real-world environments at multiple scales, ranging from objects, scenes, buildings, and ultimately entire cities. This knowledge powers intelligent tools for scene understanding which have important applications in engineering, artificial intelligence, robotics, and extended reality. Have a look at my publications for more information.

I got an MPhil in Advanced Computer Science from the University of Cambridge in 2010, before going to University College London where I got a PhD in 2015, working in the Smart Geometry Processing Group. I was supervised by Professor Niloy Mitra, working on collections of 3D objects to develop novel representations and shape analysis algorithms for modelling, learning and using object structure. I then joined the University of Cyprus for a Postdoc, where I focused on deep learning for 3D shape analysis, including semantic segmentation and style recognition.

I grew up in Cyprus, a tiny island in the Eastern Mediterranean. Have a look at my CV for more information.

news

Dec 1, 2021 Our paper Projective Urban Texturing will be presented at 3DV 2021. Check the project’s website for details including code and data.
Oct 1, 2021 Our paper BuildingNet: Learning to Label 3D Buildings will be presented as an oral paper (acceptance rate 3%) at ICCV 2021. Check the project’s website for details including code and data.
Jan 1, 2021 Our proposal to the Cyprus Research and Innovation Foundation PRE-SEED call was funded with 100k euros. More details here.

selected publications

  1. Projective Urban Texturing
    Georgiou, Yiangos,  Averkiou, Melinos, Kelly, Tom, and Kalogerakis, Evangelos
    In Proc. International Conference on 3D Vision (3DV), Dec 2021
  2. BuildingNet: Learning to Label 3D Buildings
    Selvaraju, P., Nabail, M., Loizou, M., Maslioukova, M.,  Averkiou, M., Andreou, A., Chaudhuri, S., and Kalogerakis, E.
    In Proc. IEEE/CVF International Conference on Computer Vision (ICCV), Dec 2021
  3. Learning Part Boundaries from 3D Point Clouds
    Loizou, Marios,  Averkiou, Melinos, and Kalogerakis, Evangelos
    Computer Graphics Forum (Proc. SGP), 39(5), Dec 2020
  4. Learning Material-Aware Local Descriptors for 3D Shapes
    Lin, Hubert,  Averkiou, Melinos, Kalogerakis, Evangelos, Kovacs, Balazs, Ranade, Siddhant, Kim, Vladimir, Chaudhuri, Siddhartha, and Bala, Kavita
    In Proc. International Conference on 3D Vision (3DV), Sep 2018
  5. 3D Shape Segmentation with Projective Convolutional Networks
    Kalogerakis, Evangelos,  Averkiou, Melinos, Maji, Subhransu, and Chaudhuri, Siddhartha
    In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jul 2017
  6. Co-Locating Style-Defining Elements on 3D Shapes
    Hu, Ruizhen, Li, Wenchao, Kaick, Oliver Van, Huang, Hui,  Averkiou, Melinos, Cohen-Or, Daniel, and Zhang, Hao
    ACM Transactions on Graphics, 36(3), Jun 2017
  7. Meta-Representation of Shape Families
    Fish, Noa,  Averkiou, Melinos, Kaick, Oliver, Sorkine-Hornung, Olga, Cohen-Or, Daniel, and Mitra, Niloy J.
    ACM Transactions on Graphics (Proc. SIGGRAPH), 33(4), Jul 2014