Henri Rebecq is a Senior Computer Vision Engineer with a PhD and over a decade of experience building ML and vision systems spanning deep learning, generative AI, SLAM/3D reconstruction and mixed reality. He has a strong research pedigree (h-index 20, >7,000 citations) and a track record of translating ideas into impact—leading teams at Google to ship Google Lens features used by millions and contributing industrial patents. Comfortable in both research and production, he codes in Python and C++ and has contributed to notable event-camera open-source projects (e.g., ROS drivers and the ESIM simulator) that advance high-speed, high-dynamic-range imaging. His PhD work pioneered event-camera applications for VIO, SLAM and fast video generation, and he continues to bridge robotics, AR and practical product engineering from his base in Switzerland.
Doctor of Philosophy - PhD, Computer Science, Highest Distinction (Summa Cum Laude), Doctor of Philosophy - PhD, Computer Science, Highest Distinction (Summa Cum Laude) at University of Zurich
Engineer's Degree, Computer Vision, 3D Computer Graphics, Signal Processing, Engineer's Degree, Computer Vision, 3D Computer Graphics, Signal Processing at Telecom ParisTech
Contributions:35 commits, 1 PR, 10 pushes in 10 months
Contributions summary:Henri primarily contributes to the event camera simulator by implementing and improving various aspects of the software. They added features like including R and P in camera info messages and fixed issues such as corrupted images and contrast threshold bugs. The user also updated the data provider by reading images from a folder and wrote a script to generate timestamps.
Contributions:42 commits, 25 PRs, 35 pushes in 2 years
Contributions summary:Henri focused on implementing auto-exposure functionality within the Davis ROS driver. They added dynamic reconfigurability for the exposure settings, including desired intensity. The contributions involved modifying the driver code to interact with the camera's APS, integrating with ROS message types to handle image data and control exposure, and implementing a trimmed mean for auto-exposure. They also added the ability to dynamically configure miniDAVIS346 biases, hardware filters, and IMU biases.
roboticsros2dvsslam-algorithmsros-packages
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.