Skip to main content

Rosalie TRAN

Smart Event-Based Image Processing

Event-based image sensor, Image processing, Neural networks

The PhD candidate will be in charge of developing an edge processing methodology for an event-based image sensor. The aim of this upstream study is to evaluate the interactions between the design of a new Neural Networks paradigm and the required characteristics of the frameless event-based sampling device. The idea is to conduct this study at a high level in order to guide the next step to integrate the device. He (she) will work on adapting the processing for these new cameras and compare the neural networks between standard and event-based image sensors. A first prototype design will evaluate the correct functioning and the performances of such a processing system. At longer term, a fully integrated implementation with a 3D staked image sensor is targeted.


Thesis director: Stéphane MANCINI
Thesis supervisor: Laurent FESQUET
Thesis started on: Oct. 2021
Doctoral school: MSTII

Submitted on January 12, 2022

Updated on March 17, 2022