Learning scale-variant and scale-invariant features for deep image classification
This addresses image classification challenges for computer vision researchers, but it is incremental as it builds on prior work focusing on scale-invariant representations.
The paper tackles the problem of scale variations in images hindering CNN performance by proposing a multi-scale CNN method that develops both scale-invariant and scale-variant features, resulting in improved performance over single-scale CNNs on a challenging classification task.
Convolutional Neural Networks (CNNs) require large image corpora to be trained on classification tasks. The variation in image resolutions, sizes of objects and patterns depicted, and image scales, hampers CNN training and performance, because the task-relevant information varies over spatial scales. Previous work attempting to deal with such scale variations focused on encouraging scale-invariant CNN representations. However, scale-invariant representations are incomplete representations of images, because images contain scale-variant information as well. This paper addresses the combined development of scale-invariant and scale-variant representations. We propose a multi- scale CNN method to encourage the recognition of both types of features and evaluate it on a challenging image classification task involving task-relevant characteristics at multiple scales. The results show that our multi-scale CNN outperforms single-scale CNN. This leads to the conclusion that encouraging the combined development of a scale-invariant and scale-variant representation in CNNs is beneficial to image recognition performance.