Jaakko Lehtinen
Distinguished Research Scientist, NVIDIA Research
ELLIS Fellow
Contact
Email: first.last at aalto.fi, jlehtinen at nvidia com
Mailing address: PO BOX 15400, FI-00076 Aalto, Finland
Office: B308, Aalto CS Building (Konemiehentie 2, Espoo, map)
Important resources
Important Note to prospective international students: unless accompanied by a strong recommendation from a faculty member I personally know, I will only be accepting new PhD students who I have not worked with before through the yearly call of the ELLIS PhD Program. If I'm not listed in the call, I will not be accepting new candidates.
Advice for (prospective) students
Important! Read this before contacting me on potential student exchanges etc.
Bill Freeman's excellent notes on how to do research
Frédo Durand's fun notes on writing.
Kayvon Fatahalian's advice for CMU CS students. Applies to everyone ambitous here too.
Short Bio
I am a tenured associate professor at Aalto University, and a distinguished research scientist at NVIDIA Research. I work on computer graphics, computer vision, and machine learning, with particular interests in generative modelling, realistic image synthesis, and appearance acquisition and reproduction. Overall, I'm fascinated by the combination of machine learning techniques with physical simulators in the search for robust, interpretable AI.
Prior to taking my current positions, I spent 2007-10 as a postdoc with Frédo Durand at MIT. Before my research career, I worked for the game developer Remedy Entertainment in 1996-2005 as a graphics programmer, and contributed significantly to the graphics technology behind the worldwide blockbuster hit games Max Payne (2001), Max Payne 2 (2003), and Alan Wake (2009).
In my free time, I enjoy rock climbing, photography, cooking, and music (e.g. singing).
In the Media
- New York Times interview on illusory faces generated by progressively grown GANs.
- BBC Click on progressively grown GANs.
- MIT Technology Review on progressively grown GANs.
- The Verge on progressively grown GANs.
- NVIDIA Blog on learning to restore corrupted images without ever seeing clean ones.
Suomeksi (in Finnish)
- "Kuvan ihmistä ei ole olemassa", Helsingin Sanomat on progressively grown GANs.
- Me on YLE Aamu-TV talking about progressively grown GANs.
- Suomen Kuvalehti on GANs (paywall).
- Tietoviikko on GANs.
- TEK-verkkolehti on my former student, Dr. Miika Aittala, and our work on capture and reproduction of realistic surface appearance.
Practical impact
My research has had direct practical impact in at least the following ways:
- The texture space shading (TSS) feature introduced in 2018 in NVIDIA's Turing GPU architecture has its direct roots in our 2011 work on decoupled sampling for graphics pipelines (with Jonathan Ragan-Kelley & al.).
- Our 2010 paper on sketching clothoid splines with pleasing appearance (with Ilya Baran and Jovan Popovic) is now the basis of Adobe Illustrator's freehand drawing tool.
- Our work on facial performance capture has been used extensively by Remedy Entertainment in the production of the critically-acclaimed Control (2019).
- I had a small part in the design of the NVIDIA RTX ray tracing hardware architecture introduced with the Turing GPUs in 2018.
Students, collaborators, funding
Postdocs
- Iaroslav Melekhov
PhD students
- Erik Härkönen
- Pauli Kemppinen
- Sergei Kozlukov
- Tuomas Kynkäänniemi
- Victoria Oberländer
- Heikki Timonen
Undergraduate RAs
Seyed Alireza FatemiJahromi, Dann Mensah, Emil Levo.
Former Students
- Miika Aittala (PhD 2016 → postdoc at MIT → NVIDIA Research; several thesis awards)
- Adam Celarek (Visiting MSc student → PhD at TU Wien)
- Jonathan Granskog (graphics undergrad → MSc at ETH Zürich)
- Peter Hedman (MSc 2015 → PhD at UCL; MSc thesis award)
- Markus Kettunen (PhD 2020 → postdoc at NVIDIA Research in Zurich)
- Ari Silvennoinen (at Activision Central Tech; on leave from PhD)
- Daniel Thul (Visiting MSc student 2015-16 → PhD at ETH Zürich)
Collaborators
Timo Aila, Samuli Laine, Tero Karras, Frédo Durand, Tim Weyrich, Jonathan Ragan-Kelley, Ilya Baran, Jiawen Chen, Matthias Zwicker, Marco Manzi, Fabrice Rousselle, Tzu-Mao Li, Luke Anderson.
Funding
My research is supported by the European Research Council (ERC), Academy of Finland (grant# 277833), Helsinki Doctoral Education Network in Information and Communications Technology (HICT), Aalto Media Factory, Aalto University, and NVIDIA Corporation.
Teaching
CS-C3100 Computer Graphics (Fall 2024
CS-E5520 Advanced Computer Graphics (Spring 2024)
I had the honor to be elected Teacher of the Year 2013 in the Computer Science Degree Program at Aalto, and received the Aalto University School of Science Teaching Award in 2018. Also, both my introductory and advanced computer graphics classes have been chosen by the students to be among the Best Computer Science Classes at Aalto every year I've given them. In addition, I've helped Prof. Jukka Suomela to design and teach his multiple award-winning Programming Parallel Computers class.