Computer Vision News - July 2022

BEST OF CVPR 28 Exclusive Interview Do you need machine learning and artificial intelligence for that, or is it more about aerospace engineering? Weneedboth. Of course, the start-up’s core is in aerospace engineering and robotics because we need to design a robotics satellite that can capture debris. But to be able to grab the debris, we need computer vision to analyze images and understand where the debris is and its position relative to the capture satellite so that we can grab it. I understand that someone cleaning the street is paid for by the local town, but who is paying for cleaning debris in space? Currently, the funding we have is largely from the European Space Agency (ESA). We have a mission funded by the ESA that Mathieu, you teach as well as do research. Yes, I do research in the field of machine learning for computer vision, so for image analysis and visual understanding. I teach machine learning courses at the Bachelor level and for the Master in Digital Humanities program. What do you do at ClearSpace? ClearSpace aims to develop satellites that will go into space and capture debris in orbit around the Earth. There’s a tremendous amount of debris in orbit around the Earth. It canbenon-operational satellites or actual debris resulting from collisions, and as there’s more and more of it, the chance of collisions gets higher and higher. We need to clean up the space around the Earth so that space operations remain doable. Mathieu Salzmann is a Senior Researcher at EPFL in Lausanne, Switzerland, and an artificial intelligence engineer at ClearSpace, a start-up on a mission to make space sustainable.