Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences

COLLEGE OF ENGINEERING

UC Berkeley

Transferring Visual Category Models to New Domains

Kate Saenko, Brian Kulis, Mario Fritz and Trevor Darrell

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2010-54
May 7, 2010

http://www.eecs.berkeley.edu/Pubs/TechRpts/2010/EECS-2010-54.pdf

We propose a method to perform adaptive transfer of visual category knowledge from labeled datasets acquired in one image domain to other environments. We learn a representation which minimizes the effect of shifting between source and target domains using a novel metric learning approach. The key idea of our approach to domain adaptation is to learn a metric that compensates for the transformation of the object representation that occurred due to the domain shift. In addition to being one of the first studies of domain adaptation for object recognition, this work develops a general adaptation technique that could be applied to non-image data. Another contribution is a new image database for studying the effects of visual domain shift on object recognition. We demonstrate the ability of our adaptation method to improve performance of classifiers on new domains that have very little labeled data.


BibTeX citation:

@techreport{Saenko:EECS-2010-54,
    Author = {Saenko, Kate and Kulis, Brian and Fritz, Mario and Darrell, Trevor},
    Title = {Transferring Visual Category Models to New Domains},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2010},
    Month = {May},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2010/EECS-2010-54.html},
    Number = {UCB/EECS-2010-54},
    Abstract = {We propose a method to perform adaptive transfer of visual category knowledge from labeled datasets acquired in one image domain to other environments. We learn a representation which minimizes the effect of shifting between source and target domains using a novel metric learning approach. The key idea of our approach to domain adaptation is to learn a metric that compensates for the transformation of the object representation that occurred due to the domain shift. In addition to being one of the first studies of domain adaptation for object recognition, this work develops a general adaptation technique that could be applied to non-image data.  Another contribution is a new image database for studying the effects of visual domain shift on object recognition. We demonstrate the ability of our adaptation method to improve performance of classifiers on new domains that have very little labeled data.}
}

EndNote citation:

%0 Report
%A Saenko, Kate
%A Kulis, Brian
%A Fritz, Mario
%A Darrell, Trevor
%T Transferring Visual Category Models to New Domains
%I EECS Department, University of California, Berkeley
%D 2010
%8 May 7
%@ UCB/EECS-2010-54
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/2010/EECS-2010-54.html
%F Saenko:EECS-2010-54