Talk: Collaborative XR Training Experiences for Engineering Education

Abstract:

Engineering Education is a fascinating and diverse academic landscape that is meriting increased attention as demand for highly skilled Engineers (problem-solvers) grows globally. This talk explores how the introduction of XR training experiences can help meet this demand, in part by eliminating barriers to effective educational practices, but also by making engineering courses more engaging and accessible. The talk will also highlight how the Engineering Education community is uniquely positioned to advance XR training by bringing consumers, producers, and evaluators of the technology into close proximity. As an illustration of this effect, case studies of several of such efforts will be presented, including interactive, participatory demonstrations of XR training applications that have been developed by and for the target community's students and instructors.

Talk: Training for Real

Abstract:

We are in a technological transition period. The last grand paradigm shift in terms of technology adoption, the mobile revolution, happened over a decade ago. Since then, several technologies have been competing to become the next big thing: Internet of Things, Physical Computing, Brain-Computer Interfaces, AI and Agent-based Computing, and of course Virtual and Augmented Reality. In the meantime, Machine Learning, and particularly Deep Learning now play a major role in human-computer-interaction innovations, especially ones involving knowledge about the real, physical world. In a world where technology often 'takes over' it is more important than ever that we train humans on inherently human skills and keep humanity in the technology loop. We posit that the future of learning will have to include the real world (and therefore, XR), will be highly personalized, and will need human experts, not just computing experts. And if we want to teach computers to help in this effort, we will need to customize and curate a lot of meaningful human-relevant data sets along the way.

Talk: Realtime Generation of Audio Content using Deep Learning

Abstract:

Sound simulation methods based on synthesis and propagation are frequently used to generate audio content and effects for gaming, VR and multimedia techniques. In this talk, we give an overview of our recent work on interactive sound simulation and generating real-world effects using machine learning techniques. These include realtime impact sound synthesis for arbitrary shapes, learning acoustic scattering fields of real and virtual objects for hybrid sound propagation, and scene-aware audio rendering using recordings. We describe techniques based on simulation methods to generate training datasets and use geometric deep learning methods to design the neural networks. We highlight their performance on complex benchmarks and also demonstrate their benefits to other applications, including automatic speech recognition.