Welcome ! I am a Student in Machine Learning, just graduated from the ENS (Ecole Normale Supérieure) Paris-Saclay and Ecole des Ponts ParisTech with an MSc. in Applied Mathematics, Deep Learning and Computer Vision (Master MVA).

Previously I have done various research internships in Computer Vision. The latest one was in Valeo.ai where I was supervised by Patrick Perez: my research focused on developping Driver Monitoring Sytems throuch the analysis of the driver with 3D Pose Estimation.

I am looking for a 4 to 5-6 months research internship in the field of Vision and Language/Speech models. I am open to work in any place, especially in the UK, Kroea, Japan, Northern Europ, Canada and the US.

Main Research Interests

Selected Projects

Here are some examples of the projects I have carried out, don’t hesitate to click here for more! There are more to come of course ;)

Publications Test

Domain Bridge for Unpaired Image-to-Image Translation and Unsupervised Domain Adaptation
Pizzati, F., de Charette, R., Zaccaria, M., Cerri, P.
WACV, 2020
2022

2021

Comparing Speech Models to Human Perception
Victoria Brami, Juliette Millet, Ewan Dunbar and Emmanuel Dupoux
CoML research project, March-July 2021 (Oral)

What happens in the brain when humans perceive speech? We lay the ground for a new and expansive field of research aimed at reproducing human speech perception behaviour, by developing easy-to-use reference data and evaluation tools. In short, just as the past half-century has developed and tested thousands of speech perception experiments on human listeners, we develop a set of "speech perception experiments for machines," in order to find and close the gap between human and machine.

PDF
Code
Abstract

3D Pose Estimation for Driver Monitoring
Victoria Brami, Patrick Pérez, Renaud Marlet and Souhaiel Khalfaoui.
Valeo AI research project, April-September 2022 (Oral)

Each year, almost 20 000 people die in Europe's roads in car accidents. Driver's distraction account for 20% of them. We propose to investigate the best 3D Driver Realtime Pose Estimation for action recognition purpose and understand car passengers needs. This is an opportunity for the automotive industry since driver and interior monitoring systems (DMS and IMS), which require the detailed understanding of a car’s passengers typically with a single camera, are gaining more importance every day. Systems that detect driver’s drowsiness or distraction are already deployed in numerous vehicles, and will continue to expand as new laws make them mandatory.

</td> </tr> </table>

Comparing Speech Models to Human Perception
Victoria Brami, Juliette Millet, Ewan Dunbar and Emmanuel Dupoux
CoML research project, March-July 2021 (Oral)

What happens in the brain when humans perceive speech? We lay the ground for a new and expansive field of research aimed at reproducing human speech perception behaviour, by developing easy-to-use reference data and evaluation tools. In short, just as the past half-century has developed and tested thousands of speech perception experiments on human listeners, we develop a set of "speech perception experiments for machines," in order to find and close the gap between human and machine.

</td> </tr> </table> </body>