BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Brisbane
X-LIC-LOCATION:Australia/Brisbane
BEGIN:STANDARD
DTSTART:19920301T030000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20191203T080451Z
LOCATION:Plaza Meeting Room P2
DTSTART;TZID=Australia/Brisbane:20191118T143600
DTEND;TZID=Australia/Brisbane:20191118T145700
UID:siggraphasia_SIGGRAPH Asia 2019_sess106_papers_230@linklings.com
SUMMARY:Animating Landscape: Self-Supervised Learning of Decoupled Motion 
 and Appearance for Single-Image Video Synthesis
DESCRIPTION:Technical Papers\n\nAnimating Landscape: Self-Supervised Learn
 ing of Decoupled Motion and Appearance for Single-Image Video Synthesis\n\
 nEndo, Kanamori, Kuriyama\n\nAutomatic generation of a high-quality video 
 from a single image remains a challenging task despite the recent advances
  in deep generative models. This paper proposes a method that can create a
  high-resolution, long-term animation using convolutional neural networks 
 (CNNs) from a single landscape image where we mainly focus on skies and wa
 ters. Our key observation is that the motion (e.g., moving clouds) and app
 earance (e.g., time-varying colors in the sky) in natural scenes have diff
 erent time scales. We thus learn them separately and predict them with dec
 oupled control while handling future uncertainty in both predictions by in
 troducing latent codes. Unlike previous methods that infer output frames d
 irectly, our CNNs predict spatially-smooth intermediate data, i.e., for mo
 tion, flow fields for warping, and for appearance, color transfer maps, vi
 a self-supervised learning, i.e., without explicitly-provided ground truth
 . These intermediate data are applied not to each previous output frame, b
 ut to the input image only once for each output frame. This design is cruc
 ial to alleviate error accumulation in long-term predictions, which is the
  essential problem in previous recurrent approaches. The output frames can
  be looped like cinemagraph, and also be controlled directly by specifying
  latent codes or indirectly via visual annotations. We demonstrate the eff
 ectiveness of our method through comparisons with the state-of-the-arts on
  video prediction as well as appearance manipulation.\n\nRegistration Cate
 gory: Platinum Pass, Full Conference Pass, Full Conference One-Day Pass
URL:https://sa2019.conference-program.com/presentation?id=papers_230&sess=
 sess106
END:VEVENT
END:VCALENDAR