Dieter Fox
Papers
3
Total Citations
77
H-Index
3
About
Dieter Fox is a pioneering robotics and artificial intelligence researcher whose work spans human-robot interaction, 3D scene understanding, and generalist robot learning. Based at the University of Washington and affiliated with NVIDIA Research, Fox has made sustained contributions to making robots more capable, perceptive, and accessible to everyday users. His early influential work explored how untrained individuals naturally communicate with robots through deictic gesture and language, producing foundational insights into intuitive human-robot interaction that have garnered 70 citations. This research laid critical groundwork for building robots that can understand unscripted, real-world communication rather than rigid, pre-programmed commands. More recently, Fox has turned his attention to foundation models for robotics, contributing to GR00T N1, an ambitious open foundation model designed to enable generalist humanoid robots trained on massive, diverse datasets. Complementing this, his work on 3D-MVP advances visual pretraining techniques by extending masked autoencoder approaches from 2D images into full 3D multiview representations, directly addressing the spatial understanding demands of real-world manipulation tasks. Across his career, Fox exemplifies the bridge between principled machine perception research and practical robotic deployment, making him an essential figure for students entering robotics, computer vision, or embodied AI.
Research Focus
Key Achievements
Top Papers
- 1
- 2GR00T N1: An Open Foundation Model for Generalist Humanoid Robots4 citations · 2025
- 33D-MVP: 3D Multiview Pretraining for Manipulation3 citations · 2025