S. Mohammad Mostafavi I.

AI Executive·Computer Vision·Computational Pathology·Generative AI

👋 Hello! I’m an artificial intelligence researcher specializing in computer vision and machine learning. With a passion for solving complex problems, I’ve dedicated 3+ years to medical AI, particularly in the field of computational pathology. Additionally, I have 7+ years of experience in event-based vision, focusing on event cameras. I was recently recognized as a Technical Reviewer: Gold for ICML 2026, reflecting an ongoing commitment to peer review across the major CV/ML venues. Explore my latest interests here, and/or keep reading for a concise overview about me.

Current Affiliations

SNU Postdoctoral researcher at Seoul National University Machine Perception and Reasoning lab since 2024 — collaborating on event-based vision and image-to-image translation research involving diffusion and consistency models.

DGIST KAIST In addition, affiliated with two institutes under South Korea’s Ministry of Science and ICT through the InnoCORE (Innovation + Core Researchers) program: the DGIST Bio-Embodied Physical AI Research Group Computer Vision lab and the KAIST Visual Intelligence lab.

Back to Academia after Industry

SNU Returned to academia in March 2024 at SNU’s Machine Perception and Reasoning lab — the bridge from industry (Lunit) back to research on event-based vision and generative diffusion models.

MUI-MISP Held a concurrent Visiting Researcher position at MUI-MISP in Iran (Jun 2024 – Feb 2025) — a short medical-AI residency with remote alignment to SNU.

Industry Experience after PhD

Lunit As a Senior Research Scientist and Team Leader at Lunit’s Oncology AI group, I’ve been at the forefront of driving progress. Our collaborative efforts have led to prestigious awards, impactful research papers, and even a patent. Though our team was titled model-centric, I further thrive on exploring data-centric, and engineering-centric approaches to advance the field.

Academic Journey

GIST KAIST During my Ph.D. research at GIST (and KAIST), I deeply explored many exciting applications of event cameras (event-based-vision). My work spanned areas such as image reconstruction, super-resolution, and stereo depth estimation. Recognized at top conferences, my findings have been published in esteemed journals.

My Mission My north star is enhancing business efficiency and ensuring sustainable growth in every project I undertake. I am passionate about contributing to AI for humanitarian causes, working to save lives, alleviate suffering, and uphold human dignity.

Honors & Academic Service

Technical Reviewer: Gold — International Conference on Machine Learning (ICML), 2026.
Reviewer for CVPR, ECCV, ICCV, ICML, MICCAI, IJCV, and IEEE Transactions.
Challenge Organizer — OCELOT (MICCAI 2023) and Advances in Neuromorphic Vision (ICME 2024).
1st Place, CVPR Workshop Event-based Vision Competition (2021).
Presidential Excellence Award — Best Ph.D. Dissertation, GIST (2021).

Feel free to explore my CV, GitHub and Google Scholar profiles for more details about my research and contributions.

news

May 14, 2026	Recognized as a Technical Reviewer: Gold for ICML 2026.
Feb 21, 2026	Two papers accepted to CVPR 2026: DBMSolver (Main) and FALCON (Findings).
Dec 10, 2025	Pitched at COMEUP 2025 (Korea’s flagship startup conference) representing Seoul Startup Society.
Sep 01, 2025	Starting a new posptdoctoral fellow program conducted in collaboration between KAIST Visual Intelligence Lab, DGIST Computer vision lab, and SNU Machine Perception and Reasoning Lab, this research is supported by South Korea’s Ministry of Science and ICT through the InnoCORE (Innovation + Core Researchers) program.

selected publications

CVPR 2026

DBMSolver: A Training-free Diffusion Bridge Sampler for High-Quality Image-to-Image Translation

Sankarshana Venugopal, Mohammad Mostafavi, and Jonghyun Choi

In IEEE Conf. Comput. Vis. Pattern Recog. (CVPR) – Main , 2026

👨‍💻Code Bib PDF

Code available at https://github.com/snumprlab/dbmsolver

@inproceedings{mostafavi2026dbmsolver,
  title = {DBMSolver: A Training-free Diffusion Bridge Sampler for High-Quality Image-to-Image Translation},
  author = {Venugopal, Sankarshana and Mostafavi, Mohammad and Choi, Jonghyun},
  booktitle = {IEEE Conf. Comput. Vis. Pattern Recog. (CVPR) -- Main},
  year = {2026},
}

CVPR 2026

FALCON: Fast Adaptive Lightweight Computation of Intensities and Events for Depth Estimation

Sankarshana Venugopal*, Mohammad Mostafavi*, and Jonghyun Choi

In IEEE Conf. Comput. Vis. Pattern Recog. (CVPR) – Findings , 2026

Bib PDF

@inproceedings{mostafavi2026falcon,
  title = {FALCON: Fast Adaptive Lightweight Computation of Intensities and Events for Depth Estimation},
  author = {Venugopal*, Sankarshana and Mostafavi*, Mohammad and Choi, Jonghyun},
  booktitle = {IEEE Conf. Comput. Vis. Pattern Recog. (CVPR) -- Findings},
  year = {2026},
}

CVPR 2022

Stereo depth from events cameras: Concentrate and focus on the future

Yeongwoo Nam*, Mohammad Mostafavi*, Kuk-Jin Yoon, and Jonghyun Choi

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022

👨‍💻Code Bib PDF

Code avaialable at https://github.com/yonseivnl/se-cff (37⭐)

@inproceedings{nam2022stereo,
  title = {Stereo depth from events cameras: Concentrate and focus on the future},
  author = {Nam*, Yeongwoo and Mostafavi*, Mohammad and Yoon, Kuk-Jin and Choi, Jonghyun},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages = {6114--6123},
  year = {2022},
}

🏆 J. IJCV 2021

Learning to Reconstruct HDR Images from Events, with Applications to Depth and Flow Prediction

Mohammad Mostafavi, Lin Wang, and Kuk-Jin Yoon

International Journal of Computer Vision (IJCV), 2021

Bib PDF

@article{mostafavi2021learning,
  title = {Learning to Reconstruct HDR Images from Events, with Applications to Depth and Flow Prediction},
  author = {Mostafavi, Mohammad and Wang, Lin and Yoon, Kuk-Jin},
  journal = {International Journal of Computer Vision (IJCV)},
  pages = {1--21},
  year = {2021},
  publisher = {Springer US},
}

🏆 J. TPAMI 2021

E2SRI: Learning to super-resolve intensity images from events

Mohammad Mostafavi, Yeongwoo Nam, Jonghyun Choi, and Kuk-Jin Yoon

IEEE transactions on pattern analysis and machine intelligence, 2021

Bib PDF

@article{mostafavi2021e2sri,
  title = {E2SRI: Learning to super-resolve intensity images from events},
  author = {Mostafavi, Mohammad and Nam, Yeongwoo and Choi, Jonghyun and Yoon, Kuk-Jin},
  journal = {IEEE transactions on pattern analysis and machine intelligence},
  volume = {44},
  number = {10},
  pages = {6890--6909},
  year = {2021},
  publisher = {IEEE},
}

ICCV 2021

Event-intensity stereo: Estimating depth by the best of both worlds

Mohammad Mostafavi, Kuk-Jin Yoon, and Jonghyun Choi

In Proceedings of the IEEE/CVF International Conference on Computer Vision , 2021

Bib PDF

@inproceedings{mostafavi2021event,
  title = {Event-intensity stereo: Estimating depth by the best of both worlds},
  author = {Mostafavi, Mohammad and Yoon, Kuk-Jin and Choi, Jonghyun},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
  pages = {4258--4267},
  year = {2021},
}

🏆Oral CVPR 2020

Learning to super resolve intensity images from events

Mohammad Mostafavi, Jonghyun Choi, and Kuk-Jin Yoon

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2020

👨‍💻Code Bib PDF Video

Code avaialable at https://github.com/gistvision/e2sri (50⭐)

@inproceedings{mostafavi2020learning,
  title = {Learning to super resolve intensity images from events},
  author = {Mostafavi, Mohammad and Choi, Jonghyun and Yoon, Kuk-Jin},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages = {2768--2776},
  year = {2020},
}

CVPR 2019

Event-based High Dynamic Range Image and Very High Frame Rate Video Generation using Conditional Generative Adversarial Networks

S. Mohammad Mostafavi I.*, Lin Wang*, Yo-Sung Ho, and Kuk-Jin Yoon

In IEEE Conf. Comput. Vis. Pattern Recog. (CVPR) , 2019

Bib PDF

@inproceedings{mostafavi2019event,
  title = {Event-based High Dynamic Range Image and Very High Frame Rate Video Generation using Conditional Generative Adversarial Networks},
  author = {Mostafavi I.*, S. Mohammad and Wang*, Lin and Ho, Yo-Sung and Yoon, Kuk-Jin},
  booktitle = {IEEE Conf. Comput. Vis. Pattern Recog. (CVPR)},
  year = {2019},
}