HomeStaffAssociate ProfessorsEmilian Radoi

Emilian RadoiAssoc. Professor

Picture1
areas of expertise
  • Vision-based human behaviour understanding
  • Biometrics
  • Computer Vision
education
  • PhD Computer Science – The University of Edinburgh
  • MSc Computer Science – The University of Edinburgh
  • BEng Computer Science – University Politehnica of Bucharest

Research Interests: vision-based human behaviour understanding and context awareness, and knowledge transfer across sensing modalities.

  1. Experience
    1. Associate Professor – University Politehnica of Bucharest, Jun 2023 – Present
    2. Assistant Professor (Lecturer) – University Politehnica of Bucharest, Jan 2018 – Jun 2023
    3. Teaching Assistant –University Politehnica of Bucharest, Oct 2016 – Dec 2017
    4. Founder – Medic Chat (www.medic.chat), Jan 2017 – Present
    5. Teaching Assistant – The University of Edinburgh, Sep 2015 – Jul 2016
    6. Managing Director – Digicore Systems (UK-based startup), 01.07.2013 – 28.02.2015
    7. Research Associate – The University of Edinburgh, 01.09.2011 – 29.02.2012
  2. Projects
    1. InsureAI (SMIS code/ID: 142909), Nov 2022 – Dec 2023
    2. ATLAS (17PCCDI/2018), May 2018 – Sep 2020
    3. PLANET ICT FP7 (257649), Sep 2011 – Dec 2014
    4. HOBNET ICT FP7 (257466), Sep 2011 – Jul 2013
  3. Awards
    1. ANIS Grant – Artificial Intelligence & Machine Learning (2021)
    2. Centre for Research in Computing (CRC) Research Grant (2021)
    3. Best abstract award for CSE (Complex Systems Engineering) at the SICSA PhD Conference (2014)
    4. Winner of the UKTI Sirius Programme for graduate entrepreneurs (2014)
    5. Winner of the SICSA Elevate grant to explore the opportunity to build a startup (2013)
    6. 1st Prize – Energy Micro Design Contest; International competition organised by a leading wireless chip manufacturer – Silicon Labs (2013)
    7. PhD Scholarship (2012)

Top publications

  • “The Paradox of Motion: Evidence for Spurious Correlations in Skeleton-based Gait Recognition Models”, FG 2024, IEEE.
  • “GaitPT: Gait Pyramid Transformer for Skeleton-based Gait Recognition”, FG 2024, IEEE.
  • “CrossGaze: A Strong Method for 3D Gaze Estimation in the Wild”, FG 2024, IEEE.
  • “Aligning Actions and Walking to LLM-Generated Textual Descriptions”, FG 2024, IEEE.
  • “Gait Recognition from Highly Compressed Videos”, FG 2024, IEEE.
  • “PsyMo: A Dataset for Estimating Self-Reported Psychological Traits from Gait”, WACV 2024, IEEE.
  • “Car Full View Dataset: Fine-Grained Predictions of Car Orientation from Images”, Electronics 2024.
  • “GaitMorph: Transforming Gait by Optimally Transporting Discrete Codes”, IJCB 2023, IEEE.
  • “Exploring Self-Supervised Vision Transformers for Gait Recognition in the Wild”, Sensors 2023.
  • “Learning Gait Representations with Noisy Multi-Task Learning”, Sensors 2022.
  • “Bootstrapping Road Sign Detection for Self-Driving Cars using Weakly-Supervised Learning”, RoEduNet 2022, IEEE.
  • “Image-based Fruit Recognition and Classification”, RoEduNet 2022, IEEE.
  • “Scalable Deployments for Real-Time AI Video Stream Processing”, RoEduNet 2022, IEEE.
  • “From face to gait: Weakly-supervised learning of gender information from walking patterns”, FG 2021, IEEE.
  • “Wildgait: Learning gait representations from raw surveillance streams”, Sensors 2021.
  • “A Comprehensive Survey of Indoor Localization Methods Based on Computer Vision”, Sensors 2020.
  • “Multi-Task Learning of Confounding Factors in Pose-Based Gait Recognition”, RoEduNet 2020, IEEE.
  • “Autonomous Self-Diagnosis System”, RoEduNet 2020, IEEE.
  • “Mapping Natural Language Questions to Medical Specialties”, RoEduNet 2020, IEEE.
  • “CamLoc: Pedestrian Location Estimation through Body Pose Estimation on Smart Cameras”, IPIN 2019, IEEE.
  • “Localization Systems Repository: A Platform for Open-source Localization Systems and Datasets”, IPIN 2019, IEEE.
  • “PXNOR: Perturbative Binary Neural Network”, RoEduNet 2019, IEEE.
  • “Stress Level Prediction Using Data from Wearables”, RoEduNet 2019, IEEE.
  • “Performance Evaluation of the VB-TDMA Protocol for Long-term Tracking and Monitoring of Mobile Entities in the Outdoors”, Q2SWinet 2015, ACM.
  • “Tracking and monitoring horses in the wild using wireless sensor networks”, WiMob 2015, IEEE.
  • “Prospeckz-5 – A wireless sensor platform for tracking and monitoring of wild horses”, DSD 2014, IEEE.
  • “CoAP-Mediated Hybrid Simulation and Visualisation Environment for Specknets”, PADS 2013, ACM.
  • “Evaluation of Routing Protocols for Internet-Enabled Wireless Sensor Networks”, ICWMC 2012, IARIA.

how can we help you?

Contact us at the PRECIS office or submit an inquiry online.

contacts

The core purpose of PRECIS is to enhance its infrastructure within National University for Science and Technology POLITEHNICA Bucharest, driving technological advancements and technological transfer for the development of innovative engineering products, processes and new generation technologies for a better future.
Our mission is to build a safe, supportive academic enrichment space where the engineering creators can go from ideea to product, from concept to prototype and beyond.

Ioan Stefan Sacala
President, PRECIS
Views: 8
Views: 8