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:20240214T070247Z
LOCATION:Meeting Room C4.9+C4.10\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T114000
DTEND;TZID=Australia/Melbourne:20231214T115000
UID:siggraphasia_SIGGRAPH Asia 2023_sess170_papers_473@linklings.com
SUMMARY:Diffusing Colors: Image Colorization with Text Guided Diffusion
DESCRIPTION:Technical Papers\n\nNir Zabari, Aharon Azulay, Alexey Gorkor, 
 and Tavi Halperin (Lightricks) and Ohad Fried (Reichman University)\n\nThe
  colorization of grayscale images is a complex and subjective task with si
 gnificant challenges. Despite recent progress in employing large-scale dat
 asets with deep neural networks, difficulties with controllability and vis
 ual quality persist. To tackle these issues, we present a novel image colo
 rization framework that utilizes image diffusion techniques with granular 
 text prompts.\nThis integration not only produces colorization outputs tha
 t are semantically appropriate but also greatly improves the level of cont
 rol users have over the colorization process. Our method provides a balanc
 e between automation and control, outperforming existing techniques in ter
 ms of visual quality and semantic coherence. We leverage a pretrained gene
 rative Diffusion Model, and show that we can finetune it for the colorizat
 ion task without losing its generative power or attention to text prompts.
   \nMoreover, we present a novel CLIP-based ranking model that evaluates c
 olor vividness, enabling automatic selection of the most suitable level of
  vividness based on the specific scene semantics. Our approach holds poten
 tial particularly for color enhancement and historical image colorization.
 \n\nRegistration Category: Full Access\n\nSession Chair: Xiangyu Xu (Xi'an
  Jiaotong University)
URL:https://asia.siggraph.org/2023/full-program?id=papers_473&sess=sess170
END:VEVENT
END:VCALENDAR