Transfer Learning for Illustration Classification
This work addresses the domain-specific problem of classifying artistic depictions for researchers and practitioners in computer vision, representing an incremental improvement by applying existing transfer learning techniques to a new data domain.
The paper tackles the problem of classifying illustration images, which are often ignored in image classification research, by using transfer learning from a VGG19 network pre-trained on natural images. The result is an optimized architecture that achieves 86.61% top-1 and 97.21% top-5 precision on a new dataset of labeled illustrations, while maintaining object recognition in photographs.
The field of image classification has shown an outstanding success thanks to the development of deep learning techniques. Despite the great performance obtained, most of the work has focused on natural images ignoring other domains like artistic depictions. In this paper, we use transfer learning techniques to propose a new classification network with better performance in illustration images. Starting from the deep convolutional network VGG19, pre-trained with natural images, we propose two novel models which learn object representations in the new domain. Our optimized network will learn new low-level features of the images (colours, edges, textures) while keeping the knowledge of the objects and shapes that it already learned from the ImageNet dataset. Thus, requiring much less data for the training. We propose a novel dataset of illustration images labelled by content where our optimized architecture achieves $\textbf{86.61\%}$ of top-1 and $\textbf{97.21\%}$ of top-5 precision. We additionally demonstrate that our model is still able to recognize objects in photographs.