Deep Learning & Feature Learning Methods for Vision Rob Fergus (NYU) Kai Yu (Baidu) Marc’Aurelio Ranzato (Google) Honglak Lee (Michigan) Ruslan Salakhutdinov (U. Toronto) Graham Taylor (University of Guelph) CVPR 2012 Tutorial: 9am-5:30pm
Transcript
Slide 1
Deep Learning & Feature Learning Methods for Vision Rob
Fergus (NYU) Kai Yu (Baidu) MarcAurelio Ranzato (Google) Honglak
Lee (Michigan) Ruslan Salakhutdinov (U. Toronto) Graham Taylor
(University of Guelph) CVPR 2012 Tutorial: 9am-5:30pm
Slide 2
Tutorial Overview 9.00am:IntroductionRob Fergus (NYU)
10.00am:Coffee Break 10.30am:Sparse CodingKai Yu (Baidu)
11.30am:Neural NetworksMarcAurelio Ranzato (Google) 12.30pm:Lunch
1.30pm:Restricted Boltzmann Honglak Lee (Michigan) Machines
2.30pm:Deep BoltzmannRuslan Salakhutdinov (Toronto) Machines
3.00pm:Coffee Break 3.30pm:Transfer Learning Ruslan Salakhutdinov
(Toronto) 4.00pm:Motion & Video Graham Taylor (Guelph)
5.00pm:Summary / Q & AAll 5.30pm:End
Slide 3
Overview Learning Feature Hierarchies for Vision Mainly for
recognition Many possible titles: Deep Learning Feature Learning
Unsupervised Feature Learning This talk: Basic concepts Links to
existing vision approaches
Slide 4
Existing Recognition Approach Hand- designed Feature Extraction
Trainable Classifier Image/Video Pixels Features are not learned
Trainable classifier is often generic (e.g. SVM) Object Class
Slide: Y.LeCun
Slide 5
Motivation Features are key to recent progress in recognition
Multitude of hand-designed features currently in use SIFT, HOG,
LBP, MSER, Color-SIFT. Where next? Better classifiers? Or keep
building more features? Felzenszwalb, Girshick, McAllester and
Ramanan, PAMI 2007 Yan & Huang (Winner of PASCAL 2010
classification competition)
Slide 6
What Limits Current Performance? Ablation studies on Deformable
Parts Model Felzenszwalb, Girshick, McAllester, Ramanan, PAMI10
Replace each part with humans (Amazon Turk) : Also removal of part
deformations has small (