PhD student at INRIA MIND & INRIA THOTH working on physics-informed distributed data compression using deep learning.
Recent Work
CVPR 2026
Parallelised Differentiable Straightest Geodesics for 3D Meshes
Hippolyte Verninas, Caner Korkmaz, Stefanos Zafeiriou, Tolga Birdal, Simone Foti
We propose a novel method for differentiating straightest geodesics on 3D meshes, enabling new applications in geometry processing and computer vision. Building on this approach, we develop new convolutional layers for deep learning on meshes, introduce a method for flow matching on meshes, and design a second-order optimiser tailored to mesh-based representations.
