Ronald Clark - Representation Learning for 3D Vision скачать в хорошем качестве

Ronald Clark - Representation Learning for 3D Vision

Talk abstract: For robots to understand and interact with the world around them, they need access to a good representation of their environment. In this talk we’ll look at how deep learning can help to build such a representation. We’ll look at how generative priors can be used to improve the representation learning capabilities of variational autoencoders (VAEs). We’ll also investigate how 3D geometry can help build reliable spatial memories. Finally, we’ll see how these representations can be used for accurate state estimation. In essence, representation learning for 3D vision opens up many exciting possibilities and I hope this talk will spark some interesting ideas in this direction. Speaker bio: Ronnie is a research fellow at Imperial College London. He obtained his PhD from the University of Oxford, where he held an EPSRC studentship. His research interests are in mobile perception, including robust 3D reconstruction on mobile devices, SLAM and semantic scene understanding using learning-based methods. He received a BSc and MSc degree in Information Engineering from the University of Witwatersrand, specializing in non-linear systems and control in 2014. He has received various accolades for his research including a best paper honourable mention at CVPR'18. === Robot learning seminar series webpage: http://montrealrobotics.ca/robotlearn... Schedule (Google calendar): https://calendar.google.com/calendar/... Follow us on Youtube and Twitter for more updates: Youtube channel:    / @montrealrobotics   Twitter:   / montrealrobots