Farzad Beizaee

Computer vision, Machine learning, and Medical image analysis

Resume

About Me

Ph.D. in Computer Science with over 5 years research and development experience in computer vision and machine learning, with core expertise in generative models. Strong publication record in top-tier venues like CVPR, NeurIPS, MICCAI, MedIA, etc., and passionate about pushing the boundaries of generative modeling and translating advanced machine learning into real-world solutions in multimedia generation, autonomous systems, and healthcare.

Experience

Placeholder image

Zebra Technologies

Computer vision internship - -

Developing Character Detection for OCR.

  • Modifying real-time object detection network architecture.
  • Analytic tests and validations.
  • Generating synthetic data.
pytorch
Placeholder image

BARAI startup

Computer vision scientist - -

Wide range of different AI solutions.

  • Clothes visual search and image retrieval.
  • Face authentication.
  • Multi-task attribute tagging.
  • Persian OCR.
  • Acne detection.
pytorch

VIDA startup

Computer vision scientist - -

Full face authentication.

  • Face detection and recognition.
  • spoof detection.
  • aliveness and blink detection.
pytorch
Placeholder image

Sharif University of Technology

Teaching assistantship - -

Teacher assistant in different courses in CE department.

  • Advanced 3D computer vision.
  • Deep learning.
  • Machine learning.
  • Fundamentals of programming.
pytorch

Projects

Clothes visual search and tagging

Industrial anomaly detection

Brain anomaly detection

MRI harmonization

Neonatal brain assessment

Face authentication

RGB-D Human action recognition

Optical chararcter recognition

EEG-Based driver fatigue detection

Education / Training

Placeholder image

École de technologie supérieure (ÉTS Montréal)

-

Ph.D. in enginnering.

  • Thesis: Advancing brain MRI assessment using generative models.
Placeholder image

Sharif Universiry of Technology

-

M.Sc. in computer engineering and artificial intelligence and robotics major.

  • Thesis: Human action recognition from RGB-D videos using deep neural networks.
Placeholder image

Shiraz Universiry of Technology

Electrical engineering - Control - -

B.Sc in Electrical Engineering, control major.

  • Thesis: Myocardial infarction diagnosis with ECG data.

Publication

CVPR2025

doi.org/10.48550/arXiv.2503.19357

Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detection
  • Farzad Beizaee, Gregory A. Lodygesnky, Christian Desrosiers, Jose Dolz.

doi.org/10.48550/arXiv.2503.04953

Spectral Informed Mamba for Robust Point Cloud Processing
  • Ali Bahri, Moslem Yazdanpanah, Mehrdad Noori, Sahar Dastani, Milad Cheraghalikhani, David Osowiechi, Gustavo Adolfo Vargas Hakim, Farzad Beizaee, Ismail Ben Ayed, Christian Desrosiers.

doi.org/10.48550/arXiv.2503.06369

Spectral State Space Model for Rotation-Invariant Visual Representation Learning
  • Sahar Dastani, Ali Bahri, Moslem Yazdanpanah, Mehrdad Noori, David Osowiechi, Gustavo Adolfo Vargas Hakim, Farzad Beizaee, Milad Cheraghalikhani, Arnab Kumar Mondal, Herve Lombaert, Christian Desrosiers.

MedIA

doi.org/10.1016/j.media.2025.103483

Harmonizing Flows: Leveraging normalizing flows for unsupervised and source-free MRI harmonization Medical Image Analysis.
  • Farzad Beizaee, Gregory A Lodygensky, Chris L Adamson, Deanne K Thompson, Jeanie LY Cheong, Alicia J Spittle, Peter J Anderson, Christian Desrosiers, Jose Dolz.

MICCAI2025

doi.org/10.48550/arXiv.2508.02889

REFLECT: Rectified Flows for Efficient Brain Anomaly Correction Transport.
  • Farzad Beizaee, Sina Hajimiri, Ismail Ben Ayed, Gregory A. Lodygensky, Christian Desrosiers, Jose Dolz.

IPMI2025

doi.org/10.48550/arXiv.2502.16943

MAD-AD: Masked Diffusion for Unsupervised Brain Anomaly Detection.
  • Farzad Beizaee, Gregory A. Lodygesnky, Christian Desrosiers, Jose Dolz.

IPMI2023

doi.org/10.48550/arXiv.2301.11551

Harmonizing Flows: Unsupervised MR harmonization based on normalizing flows.
  • Farzad Beizaee, Christian Desrosiers, Gregory A. Lodygesnky, Jose Dolz.

Scientific Reports

doi.org/10.1038/s41598-023-40244-z

Determining regional brain growth in premature and mature infants in relation to age at MRI using deep neural networks.
  • Farzad Beizaee, Michele Bona, Christian Desrosiers, Jose Dolz, Gregory Lodygensky.

NeurIPS2025

doi.org/10.48550/arXiv.2505.21844

Test-Time Adaptation of Vision-Language Models for Open-Vocabulary Semantic Segmentation.
  • Mehrdad Noori, David Osowiechi, Gustavo Adolfo Vargas Hakim, Ali Bahri, Moslem Yazdanpanah, Sahar Dastani, Farzad Beizaee, Ismail Ben Ayed, Christian Desrosiers

ICML2025

doi.org/10.48550/arXiv.2505.19546

SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds.
  • Ali Bahri, Moslem Yazdanpanah, Sahar Dastani, Mehrdad Noori, Gustavo Vargas Hakim, David Osowiech, Farzad Beizaee, Ismail Ben Ayed, Christian Desrosiers

Neurips2024

doi.org/10.48550/arXiv.2406.13875

WATT: Weight Average Test-Time Adaptation of CLIP.
  • David Osowiechi, Mehrdad Noori, Gustavo Vargas Hakim, Moslem Yazdanpanah, Ali Bahri, Milad Cheraghalikhani, Sahar Dastani, Farzad Beizaee, Ismail Ben Ayed, Christian Desrosiers

WACV2025

doi.org/10.48550/arXiv.2411.01116

Test-Time Adaptation in Point Clouds: Leveraging Sampling Variation with Weight Averaging.
  • Ali Bahri, Moslem Yazdanpanah, Mehrdad Noori, Sahar Dastani, Milad Cheraghalikhani, David Osowiech, Farzad Beizaee, Ismail Ben Ayed, Christian Desrosiers

Contact