Learning scale-variant and scale-invariant features for deep image classification

作者：

Highlights：

• We propose a new approach for learning scale-variant and scale-invariant features.

• The multi-scale CNN is an ensemble of scale-specific CNN.

• Our multi-scale CNN is used to achieve state-of-the-art results for artist attribution.

• The learnt deep multi-scale representation encodes both fine and coarse characteristics.

摘要

Highlights•We propose a new approach for learning scale-variant and scale-invariant features.•The multi-scale CNN is an ensemble of scale-specific CNN.•Our multi-scale CNN is used to achieve state-of-the-art results for artist attribution.•The learnt deep multi-scale representation encodes both fine and coarse characteristics.

论文关键词：Convolutional Neural Networks,Multi-scale,Artist Attribution,Scale-variant Features

论文评审过程：Received 31 January 2016, Revised 13 May 2016, Accepted 6 June 2016, Available online 9 June 2016, Version of Record 13 October 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.06.005