An analysis of the transfer learning of convolutional neural networks for artistic images
This work provides incremental insights into transfer learning effects for art analysis applications, helping researchers understand network adaptations in this domain.
The paper analyzed how convolutional neural networks adapt through transfer learning on artistic images, using visualization and quantitative metrics to show that networks specialize pre-trained filters and higher layers concentrate classes, and demonstrated that double fine-tuning with a medium-size artistic dataset improves classification on smaller datasets even when tasks change.
Transfer learning from huge natural image datasets, fine-tuning of deep neural networks and the use of the corresponding pre-trained networks have become de facto the core of art analysis applications. Nevertheless, the effects of transfer learning are still poorly understood. In this paper, we first use techniques for visualizing the network internal representations in order to provide clues to the understanding of what the network has learned on artistic images. Then, we provide a quantitative analysis of the changes introduced by the learning process thanks to metrics in both the feature and parameter spaces, as well as metrics computed on the set of maximal activation images. These analyses are performed on several variations of the transfer learning procedure. In particular, we observed that the network could specialize some pre-trained filters to the new image modality and also that higher layers tend to concentrate classes. Finally, we have shown that a double fine-tuning involving a medium-size artistic dataset can improve the classification on smaller datasets, even when the task changes.