From Aardvark to Zorro: A Benchmark of Mammal Images

Current object recognition systems aim at achieving two challenging goals: recognizing numerous object classes and learning new object classes from a small number of examples. This paper provides a benchmark for evaluating progress on these fundamental tasks. Several methods have recently proposed to utilize the commonalities between object classes in order to effectively acquire new object classes. Such methods can be termed interclass transfer techniques. However, it is currently difficult to asses which of the proposed methods maximally utilizes the shared structure of related classes. To facilitate the development, as well as the assessment of methods for dealing with multiple related classes, the proposed benchmark provides labeled images of over mammal classes. The images are organized in five levels of variability, and their labels provide information on the objects identity, location and pose. A labeled benchmark containing a large number of related classes is crucial for assessing how well proposed solutions scale up, and which methods better utilize interclass transfer.