David F. Fouhey
Papers
8
Total Citations
120
H-Index
5
About
David F. Fouhey is a computer vision and robotics researcher whose work sits at the intersection of 3D scene understanding, human-robot interaction, and embodied AI. His research tackles some of the most challenging problems in enabling machines to perceive and reason about the physical world — from interpreting 3D structure from single images to understanding how objects move in dynamic environments. Fouhey's most impactful contribution, LLM-Grounder (60 citations), demonstrates his forward-thinking integration of large language models with 3D visual grounding, equipping robots with the ability to navigate and manipulate objects using complex natural language queries. His work on 3D object interaction understanding and the Stereo4D framework further reflects his commitment to building richer spatial representations from everyday imagery. The 3D-GRAND dataset, offering millions of grounded training examples for 3D language models, underscores his focus on scaling robust, hallucination-resistant AI systems. Beyond pure computer vision, Fouhey has contributed to wearable robotics, co-authoring research on CNN-based locomotion intent recognition for assistive devices. Across his portfolio, his research consistently bridges perception and embodied intelligence, making him a notable figure for students interested in 3D vision, robotic manipulation, and the emerging frontier of spatially-aware language models.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Understanding 3D Object Interaction from a Single Image8 citations · 2023
- 4Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos6 citations · 2025
- 5
- 6
- 73D-MVP: 3D Multiview Pretraining for Manipulation3 citations · 2025
- 8