Moshiur Farazi — AI and Computer Vision researcher, Assistant Professor at UDST, Doha, Qatar
Assistant Professor, Data Science & AI
University of Doha for Science and Technology

Moshiur Farazi, Assistant Professor of AI and Computer Vision at UDST in Doha, Qatar

About

I am an Assistant Professor in the Department of Data Science and AI at the University of Doha for Science and Technology (UDST), Qatar. My research lies at the intersection of computer vision, multimodal deep learning, and vision-language models, with a particular focus on visual reasoning, visual question answering, and applied AI for real-world problems.

Previously, I was a Research Scientist at Data61–CSIRO, Australia, where I spent eight years leading cross-disciplinary AI projects spanning agriculture, marine science, fisheries monitoring, health, and space domains, in collaboration with government agencies (DSTG, AIS) and industry partners (Amazon, BlueScope). I also hold an ongoing Adjunct Faculty appointment at the Australian National University. Across Australia, Qatar, and Canada, I have collaborated and consulted with government, defence, and agricultural entities to design and deliver end-to-end AI pipelines for real-world deployment.

I completed my Ph.D. at the Australian National University (ANU) under the supervision of Nick Barnes and Salman Khan, focusing on visual reasoning and image understanding through a question-answering lens. I also hold an M.S. in Biomedical Engineering from the University of British Columbia.

Computer Vision Multimodal Deep Learning Vision-Language Models Visual Question Answering Zero-Shot Learning Object Detection Image Segmentation Applied AI

News

  • 2025 Seven new papers accepted: EMNLP 2025 (Industry Track), IJCNN 2025 (x2), DSAA 2025, and AICCSA 2025 (x4).
  • 2025 Received grant as CoPI: Life Cycle Sustainability Assessment of Hybrid Aviation Fuel Using ML-Based Optimization, USD $62,000.
  • Aug 2024 Joined University of Doha for Science and Technology (UDST) as Assistant Professor, Data Science & AI.
  • 2024 Two journal papers published: HairNet2 in Plant Methods (Springer Nature) and anomaly detection in fisheries in Fisheries Research.
  • 2023 Papers accepted at ICML 2023 (PMLR), IJCNN 2023, and DICTA 2023 (x2).
  • Jul 2022 Promoted to Research Scientist at Data61–CSIRO, leading computer vision projects with government and industry partners.
  • 2021 Paper accepted at NeurIPS 2021: Rethinking conditional GAN training with geometrically structured latent manifolds.
  • 2020 Received CVPR Outstanding Reviewer Award from the Computer Vision Foundation.

Selected Publications

Selected publications. Full list on Google Scholar.

  • 2025
    CAPSTONE: Composable Attribute-Prompted Scene Translation for Zero-Shot Vision–Language Reasoning
    Md I. Hossain, S.Z. Ridoy, M. Farazi, N. Mohammed, S. Rahman
    EMNLP 2025 — Empirical Methods in Natural Language Processing (Industry Track)
  • 2024
    HairNet2: deep learning to quantify cotton leaf hairiness, a complex genetic and environmental trait
    M. Farazi, W.C. Conaty, L. Egan, et al.
    Plant Methods, Springer Nature, 20(1)
  • 2023
    How much does initialization affect generalization?
    S. Ramasinghe, L.E. Macdonald, M. Farazi, H. Saratchandran, S. Lucey
    ICML 2023 — International Conference on Machine Learning (PMLR)
  • 2021
    Rethinking conditional GAN training: An approach using geometrically structured latent manifolds
    S. Ramasinghe, M. Farazi, S. Khan, N. Barnes, S. Gould
    NeurIPS 2021 — Conference on Neural Information Processing Systems
  • 2021
    Accuracy vs. Complexity: A Trade-off in Visual Question Answering Models
    M. Farazi, S. Khan, N. Barnes
    Pattern Recognition, Elsevier
  • 2020
    From Known to the Unknown: Transferring Knowledge to Answer Questions about Novel Visual and Semantic Concepts
    M. Farazi, S. Khan, N. Barnes
    Image and Vision Computing, Elsevier
  • 2018
    Reciprocal Attention Fusion for Visual Question Answering
    M. Farazi, S. Khan
    BMVC 2018 — British Machine Vision Conference

People

Ph.D. Students

  • Hanif Rasyidi
    Ph.D., Computer Science, ANU (Committee Chair) — Robust and Scalable Deep Learning for Historical Document Analysis
    Submitted 2026

Masters Students (Research)

  • Mohammed Al-Qassabi
    M.Sc., UDST — Architecture-Aware Temporal Segmentation for Attention-Based Video Anomaly Detection
    2025–2026
  • Mohamed Abubaker
    M.Sc., UDST — Multimodal Framework for Detecting Deceptive Websites Using Visual, Textual, and Contextual Analysis
    2025–2026
  • Muaz Albaghdadi
    M.Sc., UDST (Co-supervisor) — Deep Learning for Side-Channel Analysis on AES
    2025–2026
  • Yu Cui
    M.Phil., ANU — Visual Relationship Detection (VReBERT, ICPR 2022)
    2022
  • Ce Wang
    Research Masters, ANU — Zero-Shot Semantic Segmentation (IJCNN 2021)
    2020
  • Yizhi Zhao, Rizqi Ekoputris, Xiang Gao, Yi Guo
    Research Masters, ANU
    2021–2022

Capstone & Senior Projects

  • Privacy-Preserving Surveillance for Workplace Safety
    Capstone Group Project, UDST
    2025–2026
  • Beyond Stars: Bridging the Gap Between Ratings and Review Sentiment with LLM
    Senior Project, UDST — Published at AICCSA 2025
    2025–2026
  • Dual-Path Phishing Detection: Integrating Transformer-Based NLP with Structural URL Analysis
    Research Mentorship, UDST — Published at AICCSA 2025
    2025–2026
  • Multi-Modal Sentiment Analysis with Dynamic Attention Fusion
    Research Mentorship, UDST — Published at AICCSA 2025
    2026 (ongoing)
  • Corrosion Grading AI System
    Final Year Project, ANU (with Prof. Nick Birbilis) — Arara Kar, Joseph Ganter
    2021–2022