Learning ordered pooling weights in image classification
This work addresses a specific bottleneck in computer vision systems like CNNs and Bag-of-Words for image classification, but it is incremental as it builds on existing pooling techniques.
The paper tackled the problem of improving spatial pooling in image classification by introducing Ordered Weighted Average (OWA) aggregation operators, and the result showed that OWA-based pooling outperforms classical methods like maximum and average pooling in evaluations.
Spatial pooling is an important step in computer vision systems like Convolutional Neural Networks or the Bag-of-Words method. The spatial pooling purpose is to combine neighbouring descriptors to obtain a single descriptor for a given region (local or global). The resultant combined vector must be as discriminant as possible, in other words, must contain relevant information, while removing irrelevant and confusing details. Maximum and average are the most common aggregation functions used in the pooling step. To improve the aggregation of relevant information without degrading their discriminative power for image classification, we introduce a simple but effective scheme based on Ordered Weighted Average (OWA) aggregation operators. We present a method to learn the weights of the OWA aggregation operator in a Bag-of-Words framework and in Convolutional Neural Networks, and provide an extensive evaluation showing that OWA based pooling outperforms classical aggregation operators.