Fall 2014: Deep Learning for Computer Vision

Instructor: Abhinav Gupta
Where:     GHC 4101
When:      Tuesdays 12:00-1:20pm


Papers to Read and Discuss



Introduction - Administrative, Papers, Discussion

Abhinav Gupta


Unsupervised Learning
  • AutoEncoders [C]
    • Olsahausen and Field. Sparse Coding with an Overcomplete Basis Set: A Strategy Employed by V1? PDF
  • Restricted Boltzmann Machines [C]
    • Hinton, G. E., Osindero, S. and Teh, Y. (2006)
      A fast learning algorithm for deep belief nets. PDF
  • Quoc Le - NN [X]
    • Q. Le, M. Ranzato, R. Monga, M. Devin, K. Chen, G. Corrado, J. Dean, and A. Ng. Building high-level features using large scale unsupervised learning. In Proc. ICML, 2012. [PDF]

Carl Doersch
Xiaolong Wang

No Classes -- ECCV 2014


Supervised Learning
  • Convolutional Networks (MNIST) [I]
    • Handwritten digit recognition with a back-propagation network, Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard and L. D. Jackel (NIPS 1989) [PDF]

  • Alex NET (ImageNet Challenge) [I]
    • A. Krizhevsky, I. Sutskever, and G. E. Hinton. ImageNet classification with deep convolutional neural networks. In NIPS 2012. [PDF]
  • Visualizing Deep Networks [D]

Ishan Mishra
David Fouhey

  • Introduction to CAFFE toolbox [A]

Abhinav Shrivastava

  • Recurrent-Neural Networks [X]
  • Unsupervised + Supervised [JW]
    • Erhan, Courville, Bengio, Vincent. Why does unsupervised pre-training help deep learning?. AISTATS 2010 [PDF]

Xinlei Chen
Jacob Walker


Guest Lecture (Jia Li) -- Deep Learning @Yahoo! Labs

Jia Li (Yahoo!)

  • K. Simonyan, A. Vedaldi, and A. Zisserman. Deep inside convolutional networks: Visualising image classification models and saliency maps. Arxiv.org, 2013. [pdf]

Aayush Bansal
Jack Valmadre


Guest Lecture (Ross Girschik) -- Deep Learning @MSR            

Ross Girschik

  • OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks.
    Pierre Sermanet, David Eigen, Xiang Zhang, Michael Mathieu, Rob Fergus and Yann LeCun
    ICLR 2014 [http://arxiv.org/abs/1312.6229]
  • Dumitru Erhan, Christian Szegedy, Alexander Toshev, Dragomir Anguelov. Scalable Object Detection using Deep Neural Networks. Tech Report 2013. http://arxiv.org/abs/1312.2249

Debadeepta Dey
Naiyan Wang


  • 3D Scene Understanding
  • Domain Adaptation, Transfer
      • Deep Learning of Representations for Unsupervised and Transfer Learning, JMLR 2014 [PDF]
  • PANDA: Pose Aligned Networks for Deep Attribute Modeling. Ning Zhang, Manohar Paluri, Marc'Aurelio Ranzato, Trevor Darrell, Lubomir Bourdev. On Arxiv. http://arxiv.org/abs/1311.5591

David Fouhey
Krishna Kumar Singh


Lerrel Pinto
Gunnar Sigurdsson


Language and Vision
  • Yunchao Gong, L. Wang, M. Hodosh, J. Hockenmaier, S. Lazebnik. Improving Image-Sentence Embeddings Using Large Weakly Annotated Photo Collections. In Proceedings of the European Conference on Computer Vision (ECCV), 2014. [PDF]

Xinlei Chen


Learning Human Pose Estimation Features with Convolutional NetworksAjrun Jain, Graham W. Taylor, Christoph Bregler, Mykhaylo Andriluka, Jonathan Tompson. ICLR submission. 

DeepPose: Human Pose Estimation via Deep Neural Networks. Alexander Toshev and Christian Szegedy. To be appeared at CVPR 14. 

Varun Ramakrishnan


BUFFER, PPTs, End of Class