BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Visible and infrared image fusion
DTSTART:20190221T140000
DTEND:20190221T160000
DTSTAMP:20260527T185248Z
UID:67f80883ef390ea60f97d261292beb54ac7fb58c50dc0695abe0e4ca
CATEGORIES:Conferences - Seminars
DESCRIPTION:Fayez Lahoud\nEDIC candidacy exam\nExam president: Prof. Pasca
 l Fua\nThesis advisor: Prof. Sabine Süsstrunk\nCo-examiner: Prof. Wenzel 
 Jacob\n\nAbstract\nImage fusion is the process of joining complementary in
 formation as well as common features from a set of images. The challenge i
 n image fusion is to construct images that are appropriate\, understandabl
 e and more informative to the viewer than any of the single source images.
  Common fusion methods combine image transforms of the sources into a sing
 le fused transform from which they reconstruct a fused image.\n\nWe propos
 e to use pre-trained convolutional networks as feature extractors to compu
 te fast and cleaner fusions in comparison with the current state-of-the-ar
 t. We look at three prior works to motivate the research. First\, we prese
 nt a generic pixel-level image fusion using image gradients as a decomposi
 tion scheme. Second\, we explore visible and infrared image fusion with ne
 ural networks for pedestrian detection. Finally\, we discuss a perceptual 
 evaluation method to compare the quality of different image fusion schemes
 . Based on these works\, we propose our research plan to fuse images based
  on features extracted from pre-trained networks.\n\nBackground papers\nPe
 rceptual evaluation of different image fusion schemes\, by Alexande Toet\,
  Eric M. Franken\nSpectral Edge Image Fusion: Theory and Applications\, by
  David Connah\, Mark Samuel Drew\, Graham David Finlayson\nFully Convoluti
 onal Region Proposal Networks for Multispectral Person Detection\, by Dani
 el Konig\, Michael Adam\, Christian Javers\, Georg Layher\, Heiko Neumann\
 , Michael Teutsch\n\n\n \n\n 
LOCATION:BC 329 https://plan.epfl.ch/?room==BC%20329
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR