Presentation / Installation


Semi-supervised Reference-based Sketch Extraction Using a Contrastive Learning Framework
SessionMagical Sketching
DescriptionThis paper presents a novel multi-modal sketch extraction method that imitates individual artist styles using unpaired data in a semi-supervised manner. The proposed method overcomes limitations of existing techniques, including low-quality results and training difficulties, outperforming state-of-the-art sketch extraction and image-translation methods in both quantitative and qualitative evaluations.
Authors
Chang Wook Seo
Korea Advanced Institute of Science and Technology (KAIST)
Amirsaman Ashtari
Korea Advanced Institute of Science and Technology (KAIST)
Junyong Noh
Korea Advanced Institute of Science and Technology (KAIST)
Event Type
Technical Paper
TimeMonday, 7 August 20234:29pm - 4:40pm PDT
LocationPetree Hall C
ACM Digital Library Technical Papers pdfs
Session Time & Location
Sunday, 6 August 20236pm - 8:30pm PDTWest Hall B
Monday, 7 August 20233:45pm - 5:15pm PDTPetree Hall C
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Keywords
Image Processing
Video
Registration Categories
FC
FCS
V
VS
EFC
Next PresentationNext Presentation
AniFaceDrawing: Anime Portrait Exploration During Your Sketching