BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:0900
TZOFFSETTO:0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181210T184919Z
LOCATION:Lobby Gallery (B1F\, Glass Building)
DTSTART;TZID=Asia/Tokyo:20181205T130000
DTEND;TZID=Asia/Tokyo:20181205T140000
UID:siggraphasia_SIGGRAPH Asia 2018_sess223_pos_131@linklings.com
SUMMARY:Pre- and Post-Processes for Automatic Colorization using a Fully C
 onvolutional Network
DESCRIPTION:Poster\nFull Conference Pass (FC), Full Conference One-Day Pas
 s (1D), Basic Conference Pass (BC), Student One-Day Pass (SP), Experience 
 Pass (EP), Visitor Pass (VP), Exhibitor Pass (EP)\n\nPre- and Post-Process
 es for Automatic Colorization using a Fully Convolutional Network\n\nRamas
 samy, Kubo, Funatomi, Ishii, Maejima...\n\nAutomatic colorization is a sig
 nificant task especially for Anime industry. An original trace image to be
  colorized contains not only outlines but also boundary contour lines of s
 hadows and highlight areas. Unfortunately, these lines tend to decrease th
 e consistency among all images. Thus, this paper provides a method for a c
 leaning pre-process of anime dataset to improve the prediction quality of 
 a fully convolutional network, and a refinement post-process to enhance th
 e output of the network.
URL:https://sa2018.conference-program.com/presentation?id=pos_131&sess=ses
 s223
END:VEVENT
END:VCALENDAR