Poster
Transfer Learning by Borrowing Examples
Joseph Lim · Russ Salakhutdinov · Antonio Torralba

Mon Dec 12th 07:00 -- 11:59 PM @ None #None
in Poster Session and Reception »

Despite the recent trend of increasingly large datasets for object detection, there still exist many classes with few training examples. To overcome this lack of training data for certain classes, we propose a novel way of augmenting the training data for each class by borrowing and transforming examples from other classes. Our model learns which training instances from other classes to borrow and how to transform the borrowed examples so that they become more similar to instances from the target class. Our experimental results demonstrate that our new object detector, with borrowed and transformed examples, improves upon the current state-of-the-art detector on the challenging SUN09 object detection dataset.

Author Information

Joseph Lim (MIT)
Russ Salakhutdinov (Carnegie Mellon University)
Antonio Torralba (Massachusetts Institute of Technology)

More from the Same Authors