Please note: This seminar will be given online.
Andrea
Tagliasacchi, Research
Scientist
Google
Brain
It is not uncommon to think of computer graphics and computer vision as loosely disconnected disciplines; the former dealing with the synthesis of visual phenomena and the latter with analysis. However, recent advances in deep learning have blurred the boundary between the two. As a consequence, the research path to develop algorithms that effectively interpret the 3D scene “behind” an image has never seemed so well within reach.
In my talk, I will highlight a selection of my recent work. First, I will introduce architectures for vision that produce outputs that are readily compatible with traditional graphics toolchains. Then, I will simplify complex graphics models into representations that are straightforward to integrate within vision pipelines. The insights gained from these projects have shaped my research towards techniques for understanding visual content in a fully unsupervised fashion, for which I will showcase some recent results that exemplify my upcoming research agenda.
Bio: Andrea Tagliasacchi is a staff research scientist at Google Brain and an adjunct faculty in the computer science department at the University of Toronto. His research focuses on 3D perception, which lies at the intersection of machine learning, computer vision, and computer graphics.
In 2018, he was invited to join Google Daydream as a visiting faculty and eventually joined Google full time in 2019. Before joining Google, he was an assistant professor at the University of Victoria (2015-2017), where he held the “Industrial Research Chair in 3D Sensing.” His alma mater include EPFL (postdoc) SFU (PhD, NSERC Alexander Graham Bell fellow) and Politecnico di Milano (MSc, gold medalist).
To join this seminar on Zoom, please go to https://zoom.us/j/97704522990?pwd=V2ZBbkY1cVpGNFhEV0VlL1IzY3FEdz09.