



3DV, 2025
Project page • HAL • Bibtex
Object-level scene reconstruction with grid-based neural implicit methods as frames arrive, reusing object models across video sequences for accelerated reconstruction and higher completeness.
I am a final-year PhD student in the Willow team at Inria Paris and DI ENS. I work at the intersection of computer vision and robotics, under the supervision of Jean Ponce and Cordelia Schmid.
I am currently looking for a research position in computer vision and/or robotics following my graduation, scheduled for Summer 2025. If you are interested in my profile, please feel free to reach out to me!
Before starting my PhD, I received an engineering degree in applied mathematics and computer science from Ecole des Ponts ParisTech and a Masters degree in maths, vision and learning (MVA) from ENS Paris-Saclay. During my studies, I interned at Luckey/Airbnb in software engineering and at PhotoRoom in ML engineering.
My main interest is in building robust and efficient visual systems
that enable robots to understand the world, interact with it and
perform complex navigation and manipulation tasks.
My works, detailed below, cover robot task planning, manipulation and navigation, and 3D scene reconstruction.
Object-level scene reconstruction with grid-based neural implicit methods as frames arrive, reusing object models across video sequences for accelerated reconstruction and higher completeness.
Navigating to an instance of a given object category with vision by relying on a small spatial implicit map updated recursively.
Planning assemblies from images with STRIPS, Monte-Carlo tree search and convex optimization to handle failures in 6D pose estimation.
Inspired from Jon Barron's website.