BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Cambridge Image Analysis Seminars
SUMMARY:On the use of Gaussian models on patches for image
denoising - Antoine Houdard\, Institut de MathÃ©ma
tiques de Bordeaux
DTSTART;TZID=Europe/London:20190321T110000
DTEND;TZID=Europe/London:20190321T120000
UID:TALK121591AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/121591
DESCRIPTION:Some recent denoising methods are based on a stati
stical modeling of the image patches. In the liter
ature\, Gaussian models or Gaussian mixture models
are the most widely used priors.\nIn this present
ation\, after introducing the statistical framewor
k of patch-based image denoising\, I will propose
some clues to answer the following questions: Why
are these Gaussian priors so widely used? What inf
ormation do they encode? In the second part\, I wi
ll present a mixture model for noisy patches adapt
ed to the high dimension of the patch space. This
results in a denoising algorithm only based on sta
tistical tools\, which achieves state-of-the-art p
erformance. Finally\, I will discuss the limitatio
ns and some developments of the proposed method.
LOCATION:MR15\, Centre for Mathematical Sciences\, Wilberf
orce Road\, Cambridge
CONTACT:AI Aviles-Rivero
END:VEVENT
END:VCALENDAR