Tero Karras
Senior Distinguished Research Scientist at NVIDIA Research; primary author of the StyleGAN models
- Senior Distinguished Research Scientist, NVIDIA Research12
Tero Karras is a Senior Distinguished Research Scientist at NVIDIA Research, which Karras joined in 2009 after three years with NVIDIA's Tegra business. NVIDIA describes Karras as the primary author of the StyleGAN family of generative models and credits Karras with work on the hardware and software design of NVIDIA's RTX technology. Karras is first author of the StyleGAN, StyleGAN3, EDM, and EDM2 papers, whose official implementations NVIDIA publishes in its NVlabs GitHub organization. Karras's current research interests, per NVIDIA, are deep learning, generative models, and digital content creation.123569
Work
StyleGAN family (StyleGAN to StyleGAN3) (external site: github.com)Published research
First author of "A Style-Based Generator Architecture for Generative Adversarial Networks" (2018) and of "Alias-Free Generative Adversarial Networks" (StyleGAN3, NeurIPS 2021). The official StyleGAN3 implementation in NVlabs lists Karras first and is released under the NVIDIA Source Code License.3451
EDM and EDM2 diffusion models (external site: github.com)Published research
First author of "Elucidating the Design Space of Diffusion-Based Generative Models" (EDM, NeurIPS 2022), "Analyzing and Improving the Training Dynamics of Diffusion Models" (EDM2, CVPR 2024), and "Guiding a Diffusion Model with a Bad Version of Itself" (NeurIPS 2024). The EDM2 repository releases source code and pretrained models under CC BY-NC-SA 4.0.6789
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.