Presentation
HumanFlow: Controllable Human Image Generation via Flow Matching
DescriptionHumanFlow is a unified flow-matching framework for controllable full-body human image generation. It introduces a Control Encoder, Token-ControlNet, and topology-aware loss for structural consistency, along with the large-scale MiCoGen dataset, achieving improved controllability and fidelity over existing diffusion and flow-based methods.
Event Type
Technical Paper
TimeMonday, 20 July 20269:40am - 9:50am PDT
LocationRoom 403 B
Digital Library
PDF
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Monday, 20 July 20269:00am - 10:15am PDTRoom 403 B
Artificial Intelligence/Machine Learning
Generative AI
Full Conference Supporter
Full Conference
