photo Timothee Cour

Senior Software Engineer
Google, Mountain view
timothee dot cour at gmail
C:\tim\webpages\UPENN\webpage_UPENN\images\logo_X _texte.gif      

I work at Google, Mountain View, on computer vision and machine learning in Google Maps. Prior to that I was a research scientist at NEC-Labs (Silicon Valley) in the media analytics department,
working with Kai Yu and Yuanqing Lin. I finished my postdoc in the Willow group of INRIA / Ecole Normale Superieure, where worked with Jean Ponce and Francis Bach. I obtained my PhD in Computer Science under the supervision of Ben Taskar at the University of Pennsylvania, Philadelphia, where I also obtained my MS with Jianbo Shi. Before that I was an undergrad at the Ecole Polytechnique, France, in applied mathematics. Recent projects include: large scale image recognition, learning with weak supervision, pedestrian/face/hand detection, gesture recognition from webcam or kinect depth sensor, video understanding from images/text/audio, segmentation and graph matching. See also my (old...) research page or resume (available upon request).

Interests
Computer Vision
Machine Learning
D programming language

Videos
unsupervised face recognition in videos

Fully automatic person identification in a video. Absolutely no manual labeling was done, instead our algorithm automatically learns a person model from a video and associated screenplay, from which we extract the set of characters appearing in a scene (bottom labels). For more details, see our CVPR 2009 paper.

Toolboxes
Convex Learning from Partial Labels Toolbox new
Graph matching toolbox
Multiscale NCut Image segmentation Toolbox
Normalized Cuts Segmentation Code  

Datasets
Annotated Faces on TV Dataset new

Recent news
Our team won the ImageNet Challenge, 2010!
Machine Learning by Watching and Listening, 2009
Réf
érencer une Vidéo par Reconnaissance des Dialogues, 2009
Dans une vid
éo, les dialogues sont un mode de référencement comme un autre, 2009
Cool! Machine Learning by Watching and Listening, 2009

Publications [Google Scholar]

Shaoting Zhang, Ming Yang, Timothee Cour, Kai Yu, Dimitris Metaxas. Query Specific Fusion for Image Retrieval.  Proceedings of the European Conference on Computer Vision (ECCV), 2012.
[pdf][abstract][bib] new

Xiaoyu Wang, MingYang, Timothee Cour, Shenghuo Zhu, Kai Yu, and Tony X. Han. Contextual Weighting for Vocabulary Tree based Image Retrieval.  IEEE International Conference on Computer Vision (ICCV), 2011.
[pdf][abstract][bib] new

Timothee Cour, Ben Sapp, Ben Taskar. Learning from Partial Labels. Journal of Machine Learning Research (JMLR), 2011.
[pdf][abstract][bib]

Yuanqing Lin, Fengjun Lv, Shenghuo Zhu, Ming Yang, Timothee Cour, Kai Yu, Liangliang Cao, Thomas Huang. Large-scale image classification: fast feature extraction and SVM training. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2011.
[pdf][abstract][bib]

Timothee Cour, Ben Sapp, Akash Nagle, Ben Taskar. Talking Pictures: Temporal Grouping and Dialog-Supervised Person Recognition. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2010.
[pdf][abstract][bib] [supplement]

Thesis:
Weakly Supervised Learning from Multiple Modalities: Exploiting Video, Audio and Text for Video Understanding. University of Pennsylvania, PA, May 2009

[pdf][abstract][bib]

Timothee Cour, Ben Sapp, Chris Jordan, Ben Taskar. Learning from Ambiguously Labeled Images. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2009.
[pdf][abstract][bib] [tech report] [code] [dataset]

Timothee Cour, Chris Jordan, Eleni Miltsakaki, Ben Taskar. Movie/Script: Alignment and Parsing of Video and Text Transcription. Proceedings of the European Conference on Computer Vision (ECCV), 2008.
[pdf][abstract][bib][videos: retrieval of actions and character naming]

Timothee Cour, Ben Taskar. Video Deconstruction: Revealing narrative structure through image and text alignment. NIPS 2007 Workshop on the Grammar of Vision: Probabilistic Grammar-Based Models for Visual Scene Understanding and Object Categorization.
[pdf][abstract][bib]

Timothee Cour, Jianbo Shi. Recognizing objects by piecing together the Segmentation Puzzle.  IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2007.
[pdf][abstract][bib][poster[supplement: results on entire horse dataset (328 images)]

Timothee Cour, Jianbo Shi. Solving Markov Random Fields with Spectral Relaxation.  Artificial Intelligence and Statistics (AISTATS), 2007.

[pdf][abstract][bib]


Timothee Cour, Praveen Srinivasan, Jianbo Shi. Balanced Graph Matching. Advances in Neural Information Processing Systems (NIPS), 2006
[pdf][abstract][bib] [supplement: Affinely Constrained Rayleigh Quotients] 
[code]


Timothee Cour, Florence Benezit, Jianbo Shi. Spectral Segmentation with Multiscale Graph Decomposition. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2005.
[pdf][abstract][bib]
[code]


Timothee Cour, Nicolas Gogin, Jianbo Shi. Learning spectral graph segmentation. Artificial Intelligence and Statistics (AISTATS), 2005.
[pdf][abstract][bib][Slides]

Technical reports


Timothee Cour, Ben Sapp, Chris Jordan, Ben Taskar. Learning from Ambiguously Labeled Images. Technical Report MS-CIS-09-07, University of Pennsylvania, Philadelphia, May 2009.
[pdf][abstract]

Timothee Cour. Convex Relaxations for Markov Random Field MAP estimation. University of Pennsylvania, Philadelphia, September 2008.
[pdf]

Timothee Cour, Jianbo Shi. A learnable spectral memory graph for recognition and segmentation. Technical Report MS-CIS-04-12, University of Pennsylvania CIS Technical Reports, Philadelphia, PA, June 2004.
[pdf][abstract][bib]

Teaching

Fall 2006, Teaching Assistant: Automata, Computability, and Complexity, CSE 262 (w. Sudipto Guha)
Spring 2005, Teaching Assistant: Computer Vision, CSE 399 (w. Kostas Daniilidis)
Fall 2004, Teaching Assistant: Mathematical Foundations of Computer Science, CSE 260 (w. Sampath Kannan)



 


Last updated on April 2016