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_papers_480@linklings.com
SUMMARY:A Neural Implicit Representation for the Image Stack: Depth, All i
 n Focus, and High Dynamic Range
DESCRIPTION:Technical Papers\n\nChao Wang (Max-Planck-Institut für Informa
 tik); Ana Serrano (Universidad de Zaragoza); and Xingang Pan, Bin Chen, Ha
 ns-Peter Seidel, Karol Myszkowski, Christian Theobalt, Krzysztof Wolski, a
 nd Thomas Leimkühler (Max-Planck-Institut für Informatik)\n\nIn everyday p
 hotography, physical limitations of camera sensors and lenses frequently l
 ead to a variety of degradations in captured images such as saturation or 
 defocus blur. A common approach to overcome these limitations is to resort
  to image stack fusion, which involves capturing multiple images with diff
 erent focal distances or exposures. For instance, to obtain an all-in-focu
 s image, a set of multi-focus images is captured. Similarly, capturing mul
 tiple exposures allows for the reconstruction of high dynamic range (HDR).
 \nIn this paper, we present a novel approach that combines neural fields w
 ith an expressive camera model to achieve a unified reconstruction of an a
 ll-in-focus HDR image from an image stack. \nOur approach is composed of a
  set of specialized neural fields tailored to address specific sub-problem
 s along our pipeline:\nWe use fields to predict flow to overcome misalignm
 ents arising from lens breathing, depth and all-in-focus images to account
  for depth of field, as well as tonemapping to deal with sensor responses 
 and saturation -- all trained using a physically inspired supervision stru
 cture with a differentiable thin lens model at its core.\nAn important ben
 efit of our approach is its ability to handle these tasks simultaneously o
 r independently, providing flexible post-editing capabilities such as refo
 cusing and exposure adjustment.\nBy sampling the three primary factors in 
 photography within our framework (focal distance, aperture, and exposure t
 ime), we conduct a thorough exploration to gain valuable insights into the
 ir significance and impact on the overall image quality. \nThrough extensi
 ve validation, we demonstrate that our method outperforms existing approac
 hes in both depth-from-defocus and all-in-focus image reconstruction tasks
 . Moreover, our approach exhibits promising results in each of these three
  dimensions, showcasing its potential to enhance captured image quality an
 d provide greater control in post-processing.\n\nRegistration Category: Fu
 ll Access, Enhanced Access, Trade Exhibitor, Experience Hall Exhibitor
URL:https://asia.siggraph.org/2023/full-program?id=papers_480&sess=sess209
END:VEVENT
END:VCALENDAR