SIMONE FOTI

Currently researching diffusion models and Gaussian splatting techniques for generating high-resolution textures and materials directly on manifold surfaces with Tolga Birdal and Stefanos Zafeiriou. 

My main research goal was to investigate novel latent disentanglement methods applicable to geometric-deep-learning-based generative models. My project aimed at simplifying the generation of digital humans for movie production, virtual reality, and metaverse applications. My work found application also in plastic surgery and laparoscopic surgery of the liver

The aim of my research project at the Paris computer graphics group was to research continuous functional representations for patch-driven super-resolution of textures and materials.

The aim of my research project at the Digital Humans group was to study and implement a geometric deep learning algorithm for high-quality 3D face reconstruction that could be used in the production of Disney, Marvel, and Star Wars movies. A US patent was filed for this project.

This fully funded Master of Research was mainly focused on Machine Vision, Inverse Problem and Multi-agent Artificial Intelligence.

Developed an augmented reality GUI and researched computer vision and deep learning algorithms for the SMARTsurg European project.

I was hosted for six months in the Surgical Mechatronics Laboratory at CMU. I researched on deep learning and 3D reconstruction techniques to perform robot assisted vessel avoidance, an important part of my MSc thesis.

The customized study plan had a strong focus on electronics, biomedical image processing, artificial intelligence and deep learning. 

The program had a strong focus on computer science, mathematics and electronics.