About HCCV

The main goal of the HCCV group is to make computers observe and understand humans using commodity cameras. The main research question that drives our work is the following: How can we build systems that can observe humans through conventional cameras, build detailed 3D models of their body, hands and face, as well as recognize them, their actions, their emotions and even their intentions? While our main research direction focuses on computer vision for observing humans, we also expand it to other challenging vision problems, in which we can adopt similar methodologies. 

Our research is beneficial to a plethora of applications, such as human-computer interaction, robotics, augmented/virtual reality, video games, visual effects, cultural heritage, digital learning, digital art, computer-aided surgery, prosthetics, rehabilitation, healthcare, gesture recognition, sign language translation and reenactment, behavioural analysis/understanding, face recognition, safety, 3D printing and remote sensing. 

 

Latest News 

  • Invited talk by Antonis Argyros , Πως βλέπει η Τεχνητή Νοημοσύνη και πως τη βλέπουμε εμείς, Graduate Program on Bioethics (University of Crete/Kapodistrian University of Athens), December 16, 2025.

  • Invited talk by Antonis Argyros , Human-Centered Computer Vision: Observing and Interpreting Aspects of Human Presence, Robot Vision lecture series, Automation and Control Institute, Vienna University of Technology, Vienna, December 10, 2025.

  • Keynote talk by Antonis Argyros , Human-Centered Computer Vision: Observing and Interpreting Aspects of Human Presence, ROMANDIC Senior Research Seminar on “Advancing Robot Manipulation of Deformable Objects”, KIT, Karlsruhe, November 12-14, 2025.

  • M.Sc. defense. Michail Eleftherios Spanakis, defended successfully his M.Sc. thesis in November 2025.

  • Accepted papers at WACV'26. Two papers accepted at the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2026):
         G. Karvounas, N. Kyriazis, I. Oikonomidis, G. Pavlakos and A. Argyros, Enhancing Monocular 3D Hand Reconstruction with Learned Texture Priors.
         M. Benavent-Lledo, K. Bacharidis, V. Manousaki, K. Papoutsakis, A. Argyros and J. García-Rodríguez, Action Anticipation at a Glimpse: To What Extent Can Multimodal Cues Replace Video?.