Welcome back to Beyond the Patterns. So today I have the great pleasure to announce

Professor Dr. Matthias Niesner, who is a professor at the Technical University of Munich

and leads the Visual Computing Lab. Before, he was a visiting assistant professor at Stanford University.

Professor Niesner's research lies at the intersection of computer vision, graphics, machine learning,

and is particularly interested in cutting-edge techniques for 3D reconstruction,

semantic 3D scene understanding, video editing, and AI-driven video synthesis.

In total, he has published over 70 academic publications, including 22 papers at prestigious ACM Transactions on Graphics,

SIGGRAPH, SIGGRAPH Asia, and 24 works at the leading vision conferences CVPR, ECCV, ICCV.

Several of these works won best paper awards, including SIG CHI 2014, HPG 15, SPG 18,

and the SIGGRAPH 16 Emerging Technologies Award for the best live demo.

Professor Niesner's work enjoys wide media coverage with many articles featured in mainstream media,

including the New York Times, Wall Street Journal, Spiegel, MIT Technological Review, and many more.

And his work led to several TV appearances, such as on Jimmy Kimmel Live,

where Professor Niesner demonstrated the popular face-to-face technique.

Professor Niesner's academic YouTube channel has over 5 million views.

For his work, Professor Niesner received several awards.

He is a TUM IAS Rudolf Moosbauer Fellow since 2017.

He won the Google Faculty Award for Machine Perception in 2017,

the NVIDIA Professor Partnership Award in 2018, as well as the prestigious ERC Starting Grant in 2018,

which comes with 1.5 million euro in research funding.

In 2019, he received the EuroGraphics Young Researcher Award,

honoring the best upcoming graphics researchers in Europe.

In addition to his academic impact, Professor Niesner is a co-founder and director of Synthesia Inc.,

a brand new startup backed by Mark Cuban,

whose aim is to empower storytellers with cutting-edge AI-driven video synthesis.

So, given his young age, he has quite a few academic and industrial achievements, startups, many, many awards,

so I'm very, very much looking forward to his presentation entitled 3D Semantic Scene Understanding.

Matthias, thank you for visiting us and the stage is yours.

Thanks a lot Andreas for the kind introduction.

It's really a pleasure to be here. Well, be here is relative, I guess.

So as a Franconian, I'm always happy to come back to my roots.

I also grew up in Franconia, so this is actually my home.

And yeah, it's always fantastic to be an airline, which is sadly not the case right now,

but I hope, of course, we can continue this in the future again.

So in this talk, I decided to talk a bit about 3D Scene Understanding and 3D Scene Generation,

specifically of generative tasks with neural networks.

And I hope, I mean, of course, I'm aware that many of you are involved in medical imaging,

and of course Andreas' lab is very known for that.

But I think a lot of these techniques are actually, they can be probably adapted across domains, right?

And I think there's a good opportunity in a sense that probably sharing some expertise from both sides

can benefit each of the respective domains.

When we're talking about 3D Scene Understanding, we first, of course, have to think about how do we get even 3D data?

And there's, of course, many, many ways to get that.

And one of the things you can see here is one of the first prototypes of the Google car.

Google car is now a transition into Waymo, right?

And what these guys are doing, they have LiDAR scans, they have these fancy scanners on top of it.

And at the end of the day, they're going to get these point clouds.

And from these point clouds, they're doing things like object detection, semantic segmentation, and so on.

So originally speaking, I would say in the last decade, we have seen a lot of progress on,

generally speaking, on deep learning techniques on many domains.