Generative & multimodal AI
Generative models, visual content creation, and systems that connect language with images and video.
Principal Scientist · Amazon
Amit Agrawal
My research spans generative and multimodal AI, computer vision, and computational photography. With 20+ years connecting research, invention, and product development, I turn research advances into products used by millions of customers.
- U.S. patents
- 50+
- Peer-reviewed publications
- 55+
- Citations
- 12,000+
- h-index
- 50
Current research interests
Research overviewVisual understanding & 3D perception
Learning representations of people and scenes, from human shape and pose to 3D reconstruction.
Computational imaging
Bringing optics, geometry, and computation together to rethink how cameras capture the world.