Loick Chambon

Affiliations Sorbonne University and Valeo.ai.

I recently completed my PhD in Computer Vision at Sorbonne University and Valeo.ai, supervised by Matthieu Cord, Alexandre Boulch, and Eloi Zablocki.

My research focused on Vision Foundation Models and perception for autonomous driving, with a particular interest in feature upsampling, high-resolution visual representations, scalable 3D scene understanding, and efficient perception models for real-world applications.


💼 Open to job opportunities — feel free to reach out.


News

Feb 15, 2026 NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering accepted highlight at CVPR 2026.
Sep 13, 2025 JAFAR: Jack up Any Feature at Any Resolution accepted at NeurIPS 2025.
Sep 8, 2025 VaViM and VaVAM: Autonomous Driving through Video Generative Modeling accepted at CoRL 2025 Workshop LSRW.
Jul 7, 2025 GaussRender: Learning 3D Occupancy with Gaussian Rendering accepted at ICCV 2025.
Jun 17, 2024 PointBeV: A Sparse Approach to BeV Predictions accepted at CVPR 2024.
May 13, 2024 Towards motion forecasting with real-world perception inputs accepted at ICRA 2024.
Mar 8, 2023 PhD student at Sorbonne University and Valeo.ai.