👋 Hi, I'm Mahdiyar!
- 🧠 ML Research Scientist at Captura (since June 2025), building efficient, cutting-edge ML for imaging and creative workflows.
- 🎓 Ph.D. (ECE), Queen’s University — research in continual learning, federated learning, domain generalization, and gradient-based optimization for robust vision systems.
- 🔭 Current interests include efficient vision models, vision-language systems, debiasing, and scalable learning under distribution shift.
📚 Research & Projects
- Recent work spans vision-language debiasing, crowd counting, pedestrian detection, super-resolution, and biomedical image analysis.
- Emphasis on practical, compute-aware methods that transfer to real-world products.
📖 Publications & Peer Review
- Published in top AI/CV venues such as ICCV, AAAI, NeurIPS workshop, and ICASSP.
- Full list on Google Scholar.
- Active peer reviewer for AI and computer vision venues.
💻 Technical Skills
- Languages: Python, MATLAB, C/C++, Verilog/VHDL
- Tools/Frameworks: PyTorch, TensorFlow, OpenCV, Git, LaTeX
🌐 Connect & Collaborate
- Website: mahdiyarmm.github.io
- LinkedIn: linkedin.com/in/m-molahasani