BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium : The many ways to understand the pixels
DTSTART:20160321T101500
DTEND:20160321T113000
DTSTAMP:20260406T214604Z
UID:ed1336e7f14c22a68f546be8915c82933f57d95440d00607e375be74
CATEGORIES:Conferences - Seminars
DESCRIPTION:By : Philipp Kraehenbuehl - UC Berkeley\nIC Faculty candidateA
 bstract :\nThe field of computer vision is arguably seeing one of its most
  transformative changes in recent history. Convolutional neural networks (
 CNNs) have revolutionized the field\, reaching super-human performance on 
 some long-standing computer vision tasks\, such as image classification. T
 he success of these networks is fueled by massive amounts of human-labeled
  data. However this paradigm does not scale to a deeper and more detailed 
 understanding of images\, as it is simply too hard to collect enough human
 -labeled data. The issue is not that we humans don't understand the image\
 , but we often struggle to convey enough information to successfully super
 vise a vision system.\nIn this talk I show how computer vision can go beyo
 nd massive human supervision. This involves designing better models that d
 eal with fewer labels\, exploiting easier and more intuitive annotations\,
  or coming up with novel optimizations to train deep architectures with fa
 r fewer human annotations\, or even without any at all. I'll focus on thre
 e long standing computer vision problems: semantic segmentation\, intrinsi
 c image decomposition and dense semantic correspondences.Bio :\nPhilipp Kr
 ähenbühl is a postdoctoral researcher at UC Berkeley. He received a B.S.
  in Computer Science from ETH Zurich in 2009\, and a PhD in Computer Scien
 ce from Stanford University in 2014. Philipp's research interests lie in C
 omputer vision\, Machine learning and Computer Graphics. He is particularl
 y interested in deep learning\, efficient optimization techniques\, and st
 ructured output prediction.More information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR