homepeople › eugene bart

PROFILE:

 

Eugene Bart

Eugene's work focuses on high-level understanding and interpretation of images and has led to novel theories of human vision. He developed a system that can recognize a novel face from just a single photograph across significant variations in viewpoint, and also introduced a method of learning a novel category from a single example.

At PARC, Eugene applies computer vision and machine learning to the task of analyzing images of documents. The general goal is to enable computers to understand and interpret documents in the same ways humans can.

Dr. Evgeniy Bart received his BSc in physics and computer science from Tel Aviv University, and his MSc and PhD in computer science from the Weizmann Institute. He completed post-doctoral work at UMN and later Caltech, and also consulted for human vision scientist teams.

 

PARC publications

view publications by:  date | title | type | focus area

 

2014

Footprints: A visual search tool that supports discovery and coverage tracking

IEEE Visual Analytics Science and Technology (VAST)

9 November 2014

Multi-source anomaly detection: using across-domain and across-time peer-group consistency checks

Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications(JoWUA)

6 June 2014

2013

Multi-domain information fusion for insider threat detection

2013 IEEE Workshop on Research for Insider Threat (WRIT)

24 May 2013

2012

Invariant object recognition based on extended fragments

Frontiers in Computational Neuroscience

August 2012

Proactive insider threat detection through graph learning and psychological context

IEEE Workshop on Research for Insider Threat (WRIT)

25 May 2012

Parsing tables by probabilistic modeling of perceptual cues

10th IAPR International Workshop on Document Analysis Systems

27 March 2012

2011

Unsupervised organization of image collections: taxonomies and beyond

IEEE Transactions on Pattern Analysis and Machine Intelligence

November 2011

2010

Information extraction by finding repeated structure

International Workshop on Document Analysis Systems

9 June 2010

2009

Speeding up Gibbs sampling by variable grouping

NIPS Workshop on Applications for Topic Models: Text and Beyond

11 December 2009