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:20240214T070240Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_tog_105@linklings.com
SUMMARY:CLIP-Guided StyleGAN Inversion for Text-Driven Real Image Editing
DESCRIPTION:Technical Papers\n\nAbdul Basit Anees and Ahmet Canberk Baykal
  (Koç University), Duygu Ceylan (Adobe Research), Erkut Erdem (Hacettepe U
 niversity), and Aykut Erdem and Deniz Yuret (Koç University)\n\nResearcher
 s have recently begun exploring the use of StyleGAN-based models for real 
 image editing. One particularly interesting application is using natural l
 anguage descriptions to guide the editing process. Existing approaches for
  editing images using language either resort to instance-level latent code
  optimization or map predefined text prompts to some editing directions in
  the latent space. However, these approaches have inherent limitations. Th
 e former is not very efficient, while the latter often struggles to effect
 ively handle multi-attribute changes. To address these weaknesses, we pres
 ent CLIPInverter, a new text-driven image editing approach that is able to
  efficiently and reliably perform multi-attribute changes. The core of our
  method is the use of novel, lightweight text-conditioned adapter layers i
 ntegrated into pretrained GAN-inversion networks. We demonstrate that by c
 onditioning the initial inversion step on the Contrastive Language-Image P
 re-training (CLIP) embedding of the target description, we are able to obt
 ain more successful edit directions. Additionally, we use a CLIP-guided re
 finement step to make corrections in the resulting residual latent codes, 
 which further improves the alignment with the text prompt. Our method outp
 erforms competing approaches in terms of manipulation accuracy and photo-r
 ealism on various domains including human faces, cats, and birds, as shown
  by our qualitative and quantitative results.\n\nRegistration Category: Fu
 ll Access, Enhanced Access, Trade Exhibitor, Experience Hall Exhibitor
URL:https://asia.siggraph.org/2023/full-program?id=tog_105&sess=sess209
END:VEVENT
END:VCALENDAR