Mahsa Tajik

About Me

I am a dedicated researcher specializing in computer vision, with a keen interest in 3D reconstruction and generative models. With a solid academic background from the University of Tehran and hands-on experience from my role at FlexiTrace, I am well-equipped to advance cutting-edge research.
I am currently seeking a Ph.D. position to further my contributions to this dynamic field.

Education

• M.Sc. in Artificial Intelligence - University of Tehran
• B.Sc. in Hardware Engineering - University of Tehran
• Diploma in Mathematics and Physics - NODET (National Organization for Development of Exceptional Talents)

Experience

• Machine Learning Researcher, FlexiTrace (Oct 2023 - Present)
  Developed pose estimation models for markerless gait analysis and human pose tracking.
• Research Assistant, TaarLab (2020 - 2022)
  Developed a deep learning model for video-based person re-identification, improving accuracy by 2% with a novel attention mechanism.
• Teaching Assistant, University of Tehran (2020)
  Assisted in "Neural Networks and Deep Learning" course, designing assignments and supporting students.

Publications

• "Video-based Person Re-identification Using Attention Based on Separation Index" (2022)
  M. Tajik et al., 8th ICSPIS, IEEE. [Link]
• "Markerless Single-Camera Gait Analysis: Validation of a Novel System Against 3D Motion Capture" (2025)
  O. Safarzadeh, M. Tajik et al., Gait & Posture (accepted for GAMMA Congress 2025).

Skills