Developing Diffusion Models with 3D World Knowledge to Enable Analysis by Synthesis
Abstract
The goal of this proposal is to create Diffusion Models with 3D world knowledge which can be used to develop novel computer vision algorithms using an analysis by synthsis perspective. This involves developing DMS which are prompted by the world state W and can generate images I from a model conditioned on the world state.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Nov 09, 2024
- Source ID
- N000142412696
Entities
People
- Alan Yuille
Organizations
- Johns Hopkins University
- Office of Naval Research
- United States Navy