Doctoral Student
Email: [email protected]
Mohammad Mahdi
About me
I am Mohammadmahdi and I am joining INSAIT as a PhD student in Computer Vision with advisers Prof. Luc Van Gool and Dr. Danda Pani Paudel. My research interests are in robotics & computer vision, generative AI, and reinforcement learning. During my Master’s program at the University of Padova, I conducted research under the supervision of Prof. Lamberto Ballan. My work focused on developing RefDiff, a plug-in network designed for Stable Diffusion Models, which was specifically aimed at improving the visual grounding task using latent representations.
Education
Erasmus, University of British Columbia, Vancouver, Canada – Master of Computer Engineering – Winter Semester 2023-2024
University of Padova, Padova, Italy – MSc Computer Science – September 2021 to July 2024
Sharif University of Technology, Tehran, Iran – BSc Industrial Engineering – September 2016 to September 2020
Relevant Experience:
Computer Vision Intern at Visual Recognition Group, CTU, Prague – 2024
Teaching Assistant at Padova University – Deep Learning – 2022 to 2024
Teaching Assistant at Padova University – Big Data Computing – 2022 to 2024
Awards:
CTF CyberSecurity Olympiad, Italy – Ranked among top-20 – 2023
Best Master’s student – Department of Mathematics – Padova University – 2023
National Scientific Olympiad, Iran – Gold medal – 2021
2026
Stefan Maria Ailuro, Mario Markov, Mohammad Mahdi, Luc Van Gool, Danda Pani Paudel
More with Less: a Large Scale Remote Sensing VLM with a Simple Recipe
In: 18th Asian Conference on Computer Vision (ACCV) 2026
Carmelo Scribano, Mohammad Mahdi, Nedyalko Prisadnikov, Yuqian Fu, Giorgia Franchini, Danda Pani Paudel, Marko Bertogna, Luc Van Gool
Accelerating Vision Foundation Models with Drop-in Depthwise Convolution
In: International Conference on Pattern Recognition (ICPR 2026)
Mohammad Mahdi, Nedko Savov, Danda Pani Paudel, Luc Van GoolFrom Synchrony to Sequence: Exo-to-Ego Generation via InterpolationIn: European Conference on Computer Vision (ECCV 2026)
Mohammad Mahdi, Yuqian Fu, Nedko Savov, Jiancheng Pan, Danda Pani Paudel, Luc Van Gool
Exo2EgoSyn: Unlocking Foundation Video Generation Models for Exocentric-to-Egocentric Video Synthesis
In: European Conference on Computer Vision (ECCV 2026), Workshop on Human-Centered Multimodal Intelligence in the Wild
Jiancheng Pan, Runze Wang, Tianwen Qian, Mohammad Mahdi, Yanwei Fu, Xiangyang Xue, Xiaomeng Huang, Luc Van Gool, Danda Pani Paudel, Yuqian Fu
V^{2}-SAM: Marrying SAM2 with Multi-Prompt Experts for Cross-View Object Correspondence
In: The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026)
2025
Nedko Savov, Naser Kazemi, Mohammad Mahdi, Danda Pani Paudel, Xi Wang, Luc Van Gool
Exploration-Driven Generative Interactive Environments
In: The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2025)
