Matthias Niessner

Matthias Niessner (Matthias Nießner, born 1986) is a professor of computer science at the Technical University of Munich in Germany and head of its Visual Computing Lab, known for the real-time facial reenactment system Face2Face and 3D reconstruction research such as ScanNet; he is a co-founder of the AI video company Synthesia and in 2024 co-founded the 3D world-model company SpAItial.

Contents21 sections

Key facts

1986Birth

Matthias Nießner, often written Matthias Niessner in English, was born in 1986.⁠[1]

2010–2013Studies at FAU

He studied computer science at the University of Erlangen-Nuremberg (FAU), receiving his Diploma in 2010, and in 2013 submitted his doctoral thesis, “Subdivision Surface Rendering using Hardware Tessellation,” which received the highest honors.⁠[1]

His doctoral supervisor was Günther Greiner; he graduated in 2013.⁠[2]

2013–2017Stanford

From 2013 to 2017 he was a visiting assistant professor at Stanford University Stanford UniversityStanford University is a research university in California founded by Leland and Jane Stanford in 1885 and opened in 1891.Open the full entry, working with the Max Planck Center for Visual Computing and Communication.⁠[2]

2016Face2Face

With Justus Thies, Michael Zollhöfer, Marc Stamminger and Christian Theobalt he published Face2Face at CVPR 2016, reenacting faces in a target video in real time from a commodity webcam.⁠[3]

The paper was presented as an oral, with Nießner listed at Stanford University; outlets including Wired, The Verge and The Washington Post covered the work.⁠[4]

The Face2Face live demo won the SIGGRAPH 2016 Emerging Technologies Award.⁠[5]

He also demonstrated Face2Face on Jimmy Kimmel Live Jimmy Kimmel LiveJimmy Kimmel Live! is a late-night talk show aired by the American Broadcasting Company (ABC), hosted and executive produced by comedian Jimmy Kimmel, which premiered in 2003. In September 2025, after a monologue Kimmel delivered on the show about the killing of Charlie Kirk drew controversy, ABC suspended the show, resuming it six days later.Open the full entry.⁠[6]

2017ScanNet

With Angela Dai and others he published ScanNet at CVPR 2017, a dataset of richly annotated 3D reconstructions of indoor scenes for RGB-D scene understanding.⁠[7]

2017Joins TUM

In 2017 he became a professor at the Technical University of Munich, where he established the Visual Computing Group.⁠[8]

From that year he was a TUM-IAS Rudolph Moessbauer Fellow, and he received a Google Faculty Award for machine perception.⁠[6]

2017Co-Founds Synthesia

In 2017 he co-founded the AI video company Synthesia with Victor Riparbelli Victor RiparbelliVictor Riparbelli (born 1991) is a Danish entrepreneur who in 2017 co-founded the AI video company Synthesia in London with Steffen Tjerrild, Matthias Niessner and Lourdes Agapito and became its chief executive; he led the company from AI dubbing for film and advertising to corporate training video, and Synthesia’s valuation reached $4 billion in early 2026.Open the full entry, Steffen Tjerrild and Lourdes Agapito.⁠[9]

2018ERC Starting Grant

In 2018 he received a European Research Council Starting Grant (1.5 million euros) and an NVIDIA NVIDIANVIDIA is an American semiconductor and computing technology company that provides graphics processing units, artificial intelligence infrastructure, networking interconnects, and computing software, and is listed on Nasdaq under the ticker NVDA.Open the full entry Professor Partnership Award. Papers he co-authored won best paper awards at SIGCHI 2014, HPG 2015 and SPG 2018.⁠[10]

May 2018Deep Video Portraits

On May 29, 2018, he and co-authors including Hyeongwoo Kim and Christian Theobalt released Deep Video Portraits, which photo-realistically re-animates portrait videos using only an input video.⁠[11]

November 2018Scan2CAD

In November 2018 he released the Scan2CAD paper with TUM’s Armen Avetisyan, Angela Dai and others, on aligning CAD models with RGB-D scans.⁠[12]

With Angela Dai he also proposed Scan2Mesh, which turns incomplete range scans into structured 3D meshes.⁠[13]

2019Young Researcher Award

In 2019 Eurographics gave him its Young Researcher Award, recognizing his spatial-hashing system for real-time 3D reconstruction and Face2Face, which “set new standards” in real-time face reenactment.⁠[8]

That year he published FaceForensics++ at ICCV 2019 with Andreas Rössler, Justus Thies and others, on detecting manipulated facial images.⁠[14]

2019–2024Scan2CAD Project

His ERC project, “Scan2CAD: Learning to Digitize the Real World,” hosted by TUM, ran from January 1, 2019, to December 31, 2024, with EU funding of 1.5 million euros.⁠[15]

2022On Research and Startups

In 2022 he said his main passion remained academic research while he continued to support local startups at TUM, adding: “Research is the backbone of the economy of the future.”⁠[16]

2023Alumnus’s Startup Becomes a Unicorn

In July 2023 FAU’s Faculty of Engineering featured him, an alumnus who earned his degree and doctorate there, under the title “Unicorn status for Synthesia start up.”⁠[17]

2024Co-Founds SpAItial

In 2024 he co-founded the London-headquartered SpAItial with Luke Rogers, David Novotny and Ricardo Martin-Brualla to develop spatial foundation models.⁠[18]

May 2025Seed Round

In May 2025 SpAItial raised a $13 million seed round led by Earlybird Venture Capital, aiming to generate interactive, photorealistic 3D environments from text prompts.⁠[19]

April 2026Leave to Lead SpAItial

By April 2026 he was CEO of SpAItial and had taken leave from leading his TUM lab; he said 3D AI models are “kind of where ChatGPT for language was five years ago.”⁠[20]

May–June 2026: Echo

SpAItial’s world model Echo turns images and prompts into explorable 3D Gaussian Splat worlds; in May 2026 the company launched a developer API and Echo HQ, and in June a Startup Program.⁠[21]