Dr. Arsalan Siddique SCIT - BNU

Dr. Arslan Siddique is an Assistant Professor at the School of Computer & Information Technology (SCIT), Beaconhouse National University (BNU), Lahore. He specializes in computer vision, 3D reconstruction, and deep learning, bringing extensive research and academic experience from Europe and Russia.

His research includes developing user-assisted 3D reconstruction tools, trajectory estimation in crowded scenarios, point cloud registration, and zero-defect manufacturing using computer vision. He has also contributed to international projects and research collaborations with TeCIP, Scuola Superiore Sant’Anna, and ISTI-CNR, Pisa.

At BNU, he teaches courses in Object-Oriented Programming, Parallel & Distributed Computing, and Computer Vision, and supervises undergraduate final year projects and coursework projects related to computer vision.

• PhD in Computer Science, University of Pisa, Italy
• Master’s in Computer Science, Innopolis University, Russia 
• Bachelor’s in Electrical Engineering, University of Engineering and Technology, Lahore, Pakistan

• Assistant Professor, School of Computer & Information Technology, Beaconhouse National University (BNU), Lahore (June 2025 – Present)
  Courses Taught: CSC-213 Object-Oriented Programming, CSC-407 Parallel & Distributed Computing, CSC-340 Computer Vision
  Supervision: Undergraduate final year projects and coursework projects
• Research Fellow, TeCIP, Scuola Superiore Sant’Anna (Sept 2024 – Present)
  Projects: PURESCRAP (automatic detection/removal of impurities from steel scrap), SurfConInspect (zero-defect manufacturing)
• Graduate Research Fellow, Visual Computing Lab, ISTI-CNR, Pisa (March 2021 – Aug 2024)
  ESR 07 in EVOCATION ITN
  Developed GUI for user-assisted 3D reconstruction, user studies, and point cloud registration

• Programming: Python (Spyder, PyCharm, Jupyter Notebook), GUI development
• 3D Reconstruction: Structure-from-Motion, 3D vision, camera calibration
• Image-based 3D Modeling Tools: COLMAP, MetaShape, 3-Sweep
• Machine Learning & Deep Learning: Computer Vision applications, point cloud registration, trajectory estimation
• Tools & Platforms: Ubuntu, GitHub, Overleaf LaTeX

• Enhancing 3D reconstructions through user interactions [Thesis]
• Evaluating Image-Based Interactive 3D Modeling Tools [Paper] [Code]
• MoReLab: A software for user-assisted 3D reconstruction [Paper] [Code]
• HMRN: Heat map regression network to detect and track small objects in wide-area motion imagery [Paper] [Code]
• Evaluating deep learning methods for low-resolution point cloud registration in outdoor scenarios [Paper]
• Improved YOLOv4 for aerial object detection [Paper] [Code]
• Deep learning-based trajectory estimation of vehicles in crowded and crossroad scenarios [Paper] [Code]
• ROS-based Integration of smart space and mobile robot as Internet of Robotic Things [Paper] [Code]

• Introduction to Deep Learning with PyTorch
• Convolutional Neural Networks by deeplearning.ai
• Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization by deeplearning.ai
• Neural Networks and Deep Learning by deeplearning.ai
• STAG 2021 Conference PhD School Attendance

• Selected for the prestigious Marie Curie PhD Fellowship, University of Pisa
• Merit-based university scholarship for Master’s at Innopolis University, Russia

• Email Address: arslan.siddique@bnu.edu.pk