BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Real time Video Segmentation
DTSTART:20190829T140000
DTEND:20190829T160000
DTSTAMP:20260407T074043Z
UID:5c9fa46c62f8c75b71cff9e10c794c3ff2ba88a4843381c2d8fdc84f
CATEGORIES:Conferences - Seminars
DESCRIPTION:Evann Courdier\nEDIC candidacy exam\nExam president: Prof. Mar
 tin Jaggi\nThesis advisor: Dr. François Fleuret\nCo-examiner: Dr. Mathieu
  Salzmann\n\nAbstract\nImage semantic segmentation\, the process of assign
 ing a class label to every pixel in an image\, is a critical component for
  numerous applications from Medical Imaging to Video Surveillance to Auton
 omous Driving. Some of these applications require real-time outputs of the
  segmentation systems to be usable. However\, these systems are usually re
 latively slow and require high computing power.\nThis projects focuses on 
 bringing image segmentation to low computing power devices that need to ru
 n image segmentation real-time\, as for autonomous driving and drones. In 
 particular\, few works leverage the temporal coherence of successive image
 s inside a video to produce a more efficient network.\n\nBackground papers
 \nEncoder-Decoder  with  Atrous  Separable  Convolution  for  Semant
 ic\,Image Segmentation (DeepLabv3+)\, by Chen\, L.-C.\, et al.\nBiSeNet:Bi
 lateral Segmentation Network for Real-time Semantic Segmentation\, by Yu\,
  C.\, et al.\nLow-latency video semantic segmentation\, by Li\, Y. et al.\
 n 
LOCATION:INF 211 https://plan.epfl.ch/?room==INF%20211
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR