Learnable Pooling Regions for Image Classification

Mateusz Malinowski, Mario Fritz

Jan 17, 2013 (modified: Jan 17, 2013) ICLR 2013 conference submission readers: everyone
  • Decision: conferencePoster-iclr2013-workshop
  • Abstract: From the early HMAX model to Spatial Pyramid Matching, pooling has played an important role in visual recognition pipelines. Spatial pooling, by grouping of local codes, equips these methods with a certain degree of robustness to translation and deformation yet preserving important spatial information. Despite the predominance of this approach in current recognition systems, we have seen little progress to fully adapt the pooling strategy to the task at hand. This paper proposes a model for learning task dependent pooling scheme -- including previously proposed hand-crafted pooling schemes as a particular instantiation. In our work, we investigate the role of different regularization terms used in the proposed model together with an efficient method to train them. Our experiments show improved performance over hand-crafted pooling schemes on the CIFAR-10 and CIFAR-100 datasets -- in particular improving the state-of-the-art to 56.29% on the latter.