Presentation

<
Matting by Generation
SessionLighting and Matting With Image Generation
DescriptionThis paper redefines traditional regression-based matting as a generative modeling challenge and harnesses the capabilities of latent diffusion models enriched with extensive pre-trained knowledge to tackle this challenge. It not only produces mattes with superior resolution and detail but is also versatile and can perform both guidance-free and guidance-based matting.
Authors
Zhixiang Wang
University of Tokyo
Baiang Li
Hefei University of Technology
Jian Wang
Snap
Yu-Lun Liu
National Chiao Tung University
Jinwei Gu
Chinese University of Hong Kong
Yung-Yu Chuang
National Taiwan University
Shin'ichi Satoh
National Institute of Informatics
Event Type
Technical Paper
TimeWednesday, 31 July 202411:35am - 11:45am MDT
LocationMile High 1
ACM Digital Library Journal Papers' PDFs
Conference Papers' PDFs
Session Time & Location
Sunday, 28 July 20246:00pm - 8:45pm MDTBluebird Ballroom
Wednesday, 31 July 202410:45am - 12:15pm MDTMile High 1
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Registration Categories
Full Conference
Full Conference Supporter
Virtual Access
Exhibitor Full Conference
Wednesday
Next PresentationNext PresentationLighting and Matting With Image Generation - Interactive Discussion