Tero Karras
Senior Distinguished Research Scientist, NVIDIA | Creator of StyleGAN & Progressive GANs
Senior Distinguished Research Scientist at NVIDIA Research and one of the world's foremost pioneers in generative computer graphics and computer vision. Creator of Progressive Growing of GANs (ProGAN), StyleGAN, StyleGAN2, and StyleGAN3, setting world records for photorealistic image synthesis. Co-creator of Instant Neural Graphics Primitives (Instant NGP), revolutionizing real-time NeRF rendering and 3D reconstruction.
Areas of focus
Professional niches
Proof of Work
Instant Neural Graphics Primitives with a Multiresolution Hash Encoding (Instant NGP)
Introduced multiresolution hash encoding for neural graphics primitives, accelerating NeRF and signed distance field training from hours to seconds on a single GPU.
Hash collisions can induce high-frequency reconstruction noise in complex out-of-distribution scenes without adequate spatial bounding.
Context: Custom CUDA spatial hash tables, tiny-cuda-nn, real-time NeRF rendering.
View missionA Style-Based Generator Architecture for Generative Adversarial Networks (StyleGAN)
Invented the style-based generator architecture for GANs, introducing adaptive instance normalization (AdaIN) to separate high-level facial attributes (pose, identity) from stochastic variations (hair, freckles).
Susceptible to characteristic water-droplet artifacts in feature maps before the StyleGAN2 weight demodulation redesign.
Context: FFHQ dataset, NVIDIA DGX-1 clusters, custom CUDA kernels.
View mission