Amar Ali-bey

I eat PyTorch for lunch!

prof_pic.jpg

Montréal, QC, Canada

amar.alibey#gmail.com

Machine Learning researcher with a Ph.D. from Laval University 🇨🇦. I teach machines to understand and remember locations like humans do.

I build AI models that actually work in the real world by designing better architectures and collecting high quality data to train them.

Interests
  • AI in Robotics: Making robots see and understand where they are.
  • Multi-modal Learning: Combining different types of data (images, sensors, depth) to make smarter AI.
  • Computer Vision Tasks: Image retrieval, object detection, segmentation, tracking, and more.
  • LLM: Information retrieval for better generative models.
  • OSS: Contributing to and leading open-source initiatives.

Selected publications

  1. cvpr.jpg
    BoQ: A Place is Worth a Bag of Learnable Queries
    Amar Ali-bey, Brahim Chaib-draa, and Philippe Giguère
    In CVPR, 2024
  2. wacv.png
    MixVPR: Feature Mixing for Visual Place Recognition
    Amar Ali-bey, Brahim Chaib-draa, and Philippe Giguère
    In WACV, 2023
  3. neuro.jpeg
    GSV-Cities: Toward Appropriate Supervised Visual Place Recognition
    Amar Ali-bey, Brahim Chaib-draa, and Philippe Giguère
    Neurocomputing, 2022
  4. bmvc.png
    Global Proxy-based Hard Mining for Visual Place Recognition
    Amar Ali-Bey, Brahim Chaib-draa, and Philippe Giguere
    In BMVC, 2022