Panagiotis P. Filntisis

Postdoctoral Researcher at Athena Research Center and NTUA; Guest Researcher at MPI-IS

profile_acad.jpg

I am a Postdoctoral Researcher at the Athena Research Center and the National Technical University of Athens, supervised by Prof. Petros Maragos, and a Guest Researcher at the Max Planck Institute for Intelligent Systems. My research centers on the perception of expressive human behavior — 3D face and body modeling, affective computing, and multimodal learning — with a growing line of work on robot perception, active perception, and open-ended learning.

Some recent projects:

  • MOCHI — registration-free multi-view 3D face capture
  • SMIRK — 3D facial expression reconstruction
  • SPECTRE — visual speech-informed facial reconstruction
  • Neural Emotion Director — semantic control of facial expressions
  • VISOR — open-ended robotic exploration

news

May 5, 2026 MOCHI has been accepted to CVPR 2026, and the project page is now live!
Oct 31, 2025 Only-Style has received the ICCV 2025 P13N Workshop Best Paper Award.
Jun 15, 2025 Happy to share that our paper “A Transformer-Based Framework for Greek Sign Language Production using Extended Skeletal Motion Representations” received the PETRA 2025 Best Student Paper Award !
Sep 30, 2024 I gave an invited talk at Max Planck Institute for Intelligent Systems (Tubingen, Germany) on Advancements in 3D Facial Expression Reconstruction.
Jun 15, 2023 My PhD thesis “Multimodal Deep Learning for Emotion Recognition and Expression Synthesis with Applications in Human-Robot Interaction” has been honored with the “Iakovos Giourounlian” Award as among the best PhD theses with Industrial Interest.

selected publications

  1. MOCHI: Registration-Free Learnable Multi-View Capture of Faces in Dense Semantic Correspondence
    Panagiotis P. Filntisis, George Retsinas, Radek Danecek, Vanessa Sklyarova, Petros Maragos,  and Timo Bolkart
    CVPR 2026
  2. VGGT-HPE: Reframing Head Pose Estimation as Relative Pose Prediction
    Vasiliki Vasileiou,  Panagiotis P. Filntisis, Petros Maragos,  and Kostas Daniilidis
    CVPR-W 2026
  3. Towards Open-ended Robotic Exploration using Vision-Inspired Similarity and Foundation Models
    Panagiotis P. Filntisis, Efthymios Tsaprazlis, Paraskevas Oikonomou, Francesco Mattioli, Vieri Giuliano Santucci, George Retsinas,  and Petros Maragos
    ICRA 2025
  4. Category-Level 6D Object Pose Estimation in Agricultural Settings Using a Lattice-Deformation Framework and Diffusion-Augmented Synthetic Data
    Marios Glytsos,  Panagiotis P. Filntisis, George Retsinas,  and Petros Maragos
    IROS 2025
  5. SMIRK: 3D Facial Expressions through Analysis-by-Neural-Synthesis
    George Retsinas,  Panagiotis P. Filntisis, Radek Danecek, Victoria F. Abrevaya, Anastasios Roussos, Timo Bolkart,  and Petros Maragos
    CVPR 2024
  6. Visual Speech-Aware Perceptual 3D Facial Expression Reconstruction from Videos
    Panagiotis P. Filntisis, George Retsinas, Foivos Paraperas-Papantoniou, Athanasios Katsamanis, Anastasios Roussos,  and Petros Maragos
    CVPR-W 2023
  7. Neural Emotion Director: Speech-preserving semantic control of facial expressions in "in-the-wild" videos
    Foivos Paraperas Papantoniou,  Panagiotis P. Filntisis, Petros Maragos,  and Anastasios Roussos
    CVPR 2022
  8. An Audiovisual Child Emotion Recognition System for Child-Robot Interaction Applications
    Panagiotis P. Filntisis, Niki Efthymiou, Gerasimos Potamianos,  and Petros Maragos
    EUSIPCO 2021
  9. Emotion understanding in videos through body, context, and visual-semantic embedding loss
    Panagiotis P. Filntisis, Niki Efthymiou, Gerasimos Potamianos,  and Petros Maragos
    ECCV-W 2020
  10. Fusing body posture with facial expressions for joint recognition of affect in child–robot interaction
    Panagiotis P. Filntisis, Niki Efthymiou, Petros Koutras, Gerasimos Potamianos,  and Petros Maragos
    IEEE Robotics and Automation Letters 2019
  11. Video-realistic expressive audio-visual speech synthesis for the Greek language
    Panagiotis P. Filntisis, Athanasios Katsamanis, Pirros Tsiakoulis,  and Petros Maragos
    Speech Communication 2017