Matthias Nießner
Matthias Nießner researches visual computing and artificial intelligence. His career includes Stanford and leadership of a visual computing laboratory at TU Munich. He coauthored Face2Face, a method for transferring facial expressions into recorded video, and ScanNet, a labeled dataset for learning about three-dimensional indoor scenes. He received research awards and an ERC Starting Grant in 2018. He also cofounded Synthesia, connecting his academic work on vision and graphics with synthetic video technology.
- Works
- 271
- Citations
- 29,400
- h-index
- 71
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2016
Transfer facial expressions into video
Nießner coauthored Face2Face with Justus Thies and three other researchers. The method tracks source and target facial expressions and transfers them into a re-rendered video. The demonstration uses a live webcam stream to drive a recorded face.
-
2017
Data for understanding indoor scenes
He coauthored ScanNet, a color-and-depth video dataset with annotated camera poses, surfaces and semantic segments. The original paper describes 2.5 million views across 1,513 scenes and tests its use for several three-dimensional scene-understanding tasks.
-
2018–2019
Support for visual computing research
He received an ERC Starting Grant in 2018 and the Eurographics Young Researcher Award in 2019. His research covers three-dimensional reconstruction, scene understanding, video editing and AI-driven synthesis.
-
As of October 2026
Connect research with synthetic media
He previously worked at Stanford, led a laboratory at TU Munich and cofounded Synthesia. Synthesia develops AI-driven video synthesis.
Sources and credits
Photo: niessnerlab.org; creator not stated. License: Not stated by source. Source: niessnerlab.org. Cropped to a square and resized. Ask us to remove or replace this photo.