Machine Learning and Object Recognition 2016-2017

When: Thursdays 14h00 - 15h30
Where: ENSIMAG, room D211.
Who: Lectures by Cordelia Schmid and Jakob Verbeek.
Grading: Final grades are determined as follows
- 50% written exam,
- 50% quizes on the presented papers
- optionally paper presentation, grade for presentation can replace worst grade among quizes.
Paper presentations:
- each paper is presented by two or three students
- presentations last for 15~20 minutes, time yours in advance!
- Papers are assigned among students on first-serve basis

Class 1: September 29, 2016

Jakob Verbeek: Mixture of Gaussian clustering [slides] + classification [slides]

Cordelia Schmid: Efficient visual search [slides]
Student presentation Distinctive image features from scale-invariant keypoints
Tuan Hung VU, Saqib KHAN

Jakob Verbeek: convolutional networks for classification and dense prediction [slides]

Karteek Alahari
Student presentation: SLIC Superpixels
Resmashree KANTHARAJU, Dualet BARI, Katherine CHAO
Student presentation: Fully Convolutional Networks for Semantic Segmentation
Mahabub ALAM, Jibril FREJ
Student presentation: Deep Image Retrieval: Learning global representations for image search
Abdullah RASHEED, Matthias KOHL, Mohamed DYAB

Cordelia Schmid: Category localization [slides] + introduction action recognition [slides]

Cordelia Schmid: Optical flow [slides] + Action recognition [slides]
Student presentation: Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Shubham AGARWAL, Nitika VERMA, Erwan LE ROUX

Jakob Verbeek
Student presentation: Two-Stream Convolutional Networks for Action Recognition in Videos
Sergey RUBTSOVENKO, A. PASHEVICH
Student presentation: Convolutional Pose Machines
Ilnura USMANOVA, Joao PEREIRA, Sarah KUSHNER
Student presentation: Full flow: Optical flow estimation by global optimization over regular grids
Coline SEPPECHER, Vincent JEANSELME