BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260114T163633Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_papers_639@linklings.com
SUMMARY:Anything to Glyph: Artistic Font Synthesis via Text-to-Image Diffu
 sion Model
DESCRIPTION:ChangShuo Wang, Lei Wu, XiaoLe Liu, and Xiang Li (Shandong Uni
 versity); Lei Meng (Shandong University, Shandong Research Institute of In
 dustrial Technology); and Xiangxu Meng (Shandong University)\n\nThe automa
 tic generation of artistic fonts is a challenging task that attracts many 
 research interests. Previous methods specifically focus on glyph or textur
 e style transfer. However, we often come across creative fonts composed of
  objects in posters or logos. These fonts have proven to be a challenge fo
 r existing methods as they struggle to generate similar designs. This pape
 r proposes a novel method for generating creative artistic fonts using a p
 re-trained text-to-image diffusion model. Our model takes a shape image an
 d a prompt describing an object as input and generates an artistic glyph i
 mage consisting of such objects. Specifically, we introduce a novel heatma
 p-based weak position constraint method to guide the positioning of object
 s in the generated image, and we also propose the Latent Space Semantic Au
 gmentation Module that improves other information while constraining objec
 t position. Our approach is unique in that it can preserve the object's or
 iginal shape while constraining its position. And our training method requ
 ires only a small quantity of generated data, making it an efficient unsup
 ervised learning approach. Experimental results demonstrate that our metho
 d can generate various glyphs, including Chinese, English, Japanese, and s
 ymbols, using different objects. We also conducted qualitative and quantit
 ative comparisons with various position control methods for the diffusion 
 model. The results indicate that our approach outperforms other methods in
  terms of visual quality, innovation, and user evaluation.\n\nRegistration
  Category: Full Access, Enhanced Access, Trade Exhibitor, Experience Hall 
 Exhibitor\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_639&sess=sess209
END:VEVENT
END:VCALENDAR