Karsten Kreis

Karsten Kreis is a Principal Research Scientist at NVIDIA Research focusing on generative AI.

Qinsheng Zhang

Yunhao Ge

Generalizable One-shot 3D Neural Head Avatar

We present a method that reconstructs and animates a 3D head avatar from a single-view portrait image. Existing methods either involve time-consuming optimization for a specific person with multiple images, or they struggle to synthesize intricate appearance details beyond the facial region. To address these limitations, we propose a framework that not only generalizes to unseen identities based on a single-view image without requiring person-specific optimization, but also captures characteristic details within and beyond the face area (e.g. hairstyle, accessories, etc.).

Convolutional State Space Models for Long-Range Spatiotemporal Modeling

Effectively modeling long spatiotemporal sequences is challenging due to the need to model complex spatial correlations and long-range temporal dependencies simultaneously. ConvLSTMs attempt to address this by updating tensor-valued states with recurrent neural networks, but their sequential computation makes them slow to train. In contrast, Transformers can process an entire spatiotemporal sequence, compressed into tokens, in parallel. However, the cost of attention scales quadratically in length, limiting their scalability to longer sequences.

Hanzi Mao

Huck Yang

I am a Sr. Research Scientist at NV Research. 

I obtained my Ph.D. and M.Sc. from Georgia Institute of Technology, USA with Wallace H. Coulter fellowship and my B.Sc. from National Taiwan University. 

My primary research lies in the area of Multilingual Model Alignments and Speech-Language Modeling. Specifically:

Dale Durran

Durran has a 25% appointment as a Principal Research Scientist in Climate Modeling at NVIDIA and a 60% appointment as a Professor of Atmospheric Sciences at the University of Washington.  At NVIDIA his research focus in on deep learning earth-system modeling for sub-seasonal and seasonal forecasting, forecast ensembles, and generative methods for  fine-scale modeling of convective precipitation and other mesoscale fields.

David W Romero

I am a Research Scientist in Efficient Generative AI at NVIDIA’s Deep Imagination Research Team, and a finishing PhD candidate at the Vrije Universiteit Amsterdam. My research interests include all aspects of efficiency in Deep Learning, particularly the computational and parameter efficiency of long-context models and their application to generative models.

Yifan Ding

Subscribe to