Hughes Perreault - AI scientist
About
Hi! I'm Hughes, a computer vision expert specializing in object recognition and general scene
understanding in urban environments. My main focus is using computer vision technology to
increase safety and automation in transportation applications. I have extensive experience with
deep learning tools such as TensorFlow and PyTorch, as well as classic computer vision. During
my Ph.D., I achieved noteworthy results on a variety of road user datasets, such as Cityscapes,
KITTI, India Driving, UA-DETRAC, and UAVDT. Currently, I'm an AI scientist at Thales, where I
focus on optimization, multimodal image registration and fusion, object detection, and tracking.
Fun fact, my Dijkstra number is five (click here) and my Erdos number is also five (click
here).
Experiences
AI Scientist
October 2021 - Present
Thales Group
Lecturer for the INF1005C course
Trimester F19
Polytechnique Montréal
TA for the INF8770 course
Trimesters F17, W18, F18, W19, F20, W21
Polytechnique Montréal
Computer Vision Intern
August 2016 - December 2016
Genetec
C++ programmer/Technical Support
October 2014 - June 2015
Dash Computer Solutions
For specific references, please contact me.
Publications
2023
- A Lightweight Convolutional Neural Network for Target Detection in an Edge Environment
Hughes Perreault, Ola Ahmad, Daniel Czuboka
2023 Ground Vehicle Systems Engineering and Technology
Symposium (GVSETS)
- MobileFuse: Multimodal Image Fusion at the Edge
Hughes Perreault, Benoit Debaque, Rares David, Marc-Antoine
Drouin, Nicolas Duclos-Hindie, Simon Roy
2023 26th International Conference on Information Fusion
(FUSION)
2022
- Multimodal Deep Homography Estimation Using a Domain Adaptation Generative Adversarial
Network
Thomas Pouplin, Hughes Perreault, Benoit Debaque,
Marc-Antoine Drouin, Nicolas Duclos-Hindie, Simon Roy
2nd Workshop on Multimodal Big Data (MMBD 2022)
- Thermal and Visible Image Registration Using Deep Homography
Benoit Debaque, Hughes Perreault, Jean-Philippe
Mercier, Marc-Antoine Drouin, Rares David, Benedicte Chatelais, Nicolas
Duclos-Hindie, Simon Roy
2022 25th International Conference on Information
Fusion (FUSION)
2021
- Détection et segmentation des usagers de la route dans des images et vidéos
Hughes Perreault
Ph.D. thesis
- PolyTrack: Tracking with Bounding Polygons
Gaspar Faure, Hughes Perreault, Guillaume-Alexandre
Bilodeau, Nicolas Saunier
International Conference on Image Analysis and
Recognition
- Centerpoly: Real-time instance segmentation using bounding polygons
Hughes Perreault, Guillaume-Alexandre Bilodeau, Nicolas
Saunier, Maguelonne Héritier
Proceedings of the IEEE/CVF International Conference on
Computer Vision
- FFAVOD: Feature fusion architecture for video object detection
Hughes Perreault, Guillaume-Alexandre Bilodeau, Nicolas
Saunier, Maguelonne Héritier
Pattern Recognition Letters
2020
- SpotNet: Self-attention multi-task network for object detection
Hughes Perreault, Guillaume-Alexandre Bilodeau, Nicolas
Saunier, Maguelonne Héritier
17th Conference on Computer and Robot Vision (CRV)
- RN-VID: A Feature Fusion Architecture for Video Object Detection
Hughes Perreault, Maguelonne Heritier, Pierre Gravel,
Guillaume-Alexandre Bilodeau, Nicolas Saunier
International Conference on Image Analysis and
Recognition
2019
- Road user detection in videos
Hughes Perreault, Guillaume-Alexandre Bilodeau, Nicolas
Saunier, Pierre Gravel