Learnable Pooling Regions for Image Classification
By Mateusz Malinowski, Mario Fritz. 2013
40 citations recorded by the source corpus.
Biologically inspired, 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 showing that the smooth regularization term is crucial to achieve strong performance using the presented architecture. Finally, we propose an efficient and parallel method to train the model. 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.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mateusz Malinowski
- Mario Fritz
- Thomas Brox
- Hamed Pirsiavash
- Simon Ging
- Mohammadreza Zolfaghari
- Trevor Darrell
- Bernt Schiele
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Trevor Darrell
- Jitendra Malik
- Dacheng Tao
- Bernt Schiele
- Andrew Zisserman
- Xiaogang Wang
- Li Fei-Fei
Citations in the corpus
- Network In Network (2013, neutral): “Method Test Error Learned Pooling [16] 43.”
- Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review (2017, unknown): “Additionally, when tested on the CIFAR-10 and CIFAR-100 benchmarks (Krizhevsky, 2009), it also provided enhanced classification performance over average, max, and stochastic pooling (Zeiler & Fergus, 2013), and the parameterized scheme introduced by Malinowski and Fritz (2013).”
- COOT: Cooperative Hierarchical Transformer for Video-Text Representation Learning (2020, background): “There have been some attempts in other domains to design a context-aware feature fusion method [30, 31, 32, 33].”
- Learning Smooth Pooling Regions for Visual Recognition (2013, methodology): “Moreover, we extend the learnable pooling regions [23] to the events recognition task with object banks as high level features.”