Amirhossein Dadashzadeh

Amir.jpeg

a.dadashzadeh@bristol.ac.uk

Hi! I’m a Research Associate in Computer Vision at the University of Bristol, where I work on the TORUS project, developing AI-based video systems to monitor Parkinson’s patients’ movement in home environments over extended periods. I completed my PhD in Computer Vision at University of Bristol, where I focused on deep learning strategies for Parkinson’s disease assessment via video data. My research was supervised by Professor Majid Mirmehdi and Professor Alan Whone.

I am interested in video understanding, and fascinated by how we can make learning systems more adaptive and efficient, particularly under limited or unlabeled data settings, changing environments, and practical constraints.

news

May 15, 2025 New on arXiv!“Co-STAR: Collaborative Curriculum Self-Training with Adaptive Regularization for Source-Free Video Domain Adaptation” is now available.
📄 Read on arXiv – Code coming soon!
Aug 15, 2024 SEA accepted at NeurIPS 2024!
Paper on reducing temporal rollout error in long-sequence PDE generation.
🧠 Code available here → GitHub Repository
Dec 01, 2023 PFED5 released! – A Parkinson’s disease facial expression dataset with 41 patients, 5 expressions, and MDS-UPDRS scores.
➡️ Download PFED5
Oct 17, 2023 PECoP @ WACV 2024! – Code and PD4T dataset now available. ➡️ GitHub

selected publications

  1. co-star.png
    Co-STAR: Collaborative Curriculum Self-Training with Adaptive Regularization for Source-Free Video Domain Adaptation
    Amirhossein Dadashzadeh, Parsa Esmati, and Majid Mirmehdi
    arXiv preprint arXiv:2504.11669, 2025
  2. trajec.png
    Trajectory-guided Motion Perception for Facial Expression Quality Assessment in Neurological Disorders
    Shuchao Duan, Amirhossein Dadashzadeh, Alan Whone, and 1 more author
    In IEEE International Conference on Automatic Face and Gesture Recognition (FG), 2025
    Accepted
  3. SEA.png
    SEA: State-Exchange Attention for High-Fidelity Physics Based Transformers
    Parsa Esmati, Amirhossein Dadashzadeh, Vahid Ardakani, and 2 more authors
    In Advances in Neural Information Processing Systems, 2024
  4. pecop.png
    Pecop: Parameter efficient continual pretraining for action quality assessment
    Amirhossein Dadashzadeh, Shuchao Duan, Alan Whone, and 1 more author
    In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2024
  5. aug.png
    Auxiliary learning for self-supervised video representation via similarity-based knowledge distillation
    Amirhossein Dadashzadeh, Alan Whone, and Majid Mirmehdi
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022