 |
    |
Which Looks Like Which: Exploring Inter-class Relationships in Fine-Grained Visual Categorization
Jian Pu1, Yu-Gang Jiang1, Jun Wang2, and Xiangyang Xue1
1School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, Shanghai, China
jianpu@fudan.edu.cn
ygj@fudan.edu.cn
xyxue@fudan.edu.cn
2IBM T. J. Watson Research Center, Yorktown Heights, NY, USA
wangjun@us.ibm.com
Abstract. Fine-grained visual categorization aims at classifying visual data at a subordinate level, e.g., identifying different species of birds. It is a highly challenging topic receiving significant research attention recently. Most existing works focused on the design of more discriminative feature representations to capture the subtle visual differences among categories. Very limited efforts were spent on the design of robust model learning algorithms. In this paper, we treat the training of each category classifier as a single learning task, and formulate a generic multiple task learning (MTL) framework to train multiple classifiers simultaneously. Different from the existing MTL methods, the proposed generic MTL algorithm enforces no structure assumptions and thus is more flexible in handling complex inter-class relationships. In particular, it is able to automatically discover both clusters of similar categories and outliers. We show that the objective of our generic MTL formulation can be solved using an iterative reweighted 
2 method. Through an extensive experimental validation, we demonstrate that our method outperforms several state-of-the-art approaches.
Keywords: Fine-grained visual categorization, inter-class relationship, multiple task learning
LNCS 8691, p. 425 ff.
lncs@springer.com
© Springer International Publishing Switzerland 2014