Presentation / Installation
Sketch-guided Text-to-image Diffusion Models
SessionText-Guided Generation
DescriptionThis work introduces a universal approach to guide a pretrained text-to-image diffusion model with a spatial conditioning. It is data efficient, avoids large-scale training of a dedicated guidance model, and requires as little as a one-hour training on a single GPU.
Event Type
Technical Paper
TimeThursday, 10 August 20239:21am - 9:31am PDT
LocationPetree Hall D
ACM Digital Library
Technical Paper PDF
Session Time & Location
Research & Education
Livestreamed
Recorded
Artificial Intelligence/Machine Learning
Video
FC
FCS
V
VS
EFC