Selective Style Transfer for Text
This work addresses the challenge of enhancing text datasets for improved detection in computer vision, though it is incremental in applying existing style transfer concepts to text.
The paper tackled the problem of applying image style transfer to text while preserving transcriptions, demonstrating feasibility across text domains and proposing two architectures for selective style transfer. It showed that using this technique for data augmentation in scene text detection datasets boosted detector performance, with concrete improvements noted.
This paper explores the possibilities of image style transfer applied to text maintaining the original transcriptions. Results on different text domains (scene text, machine printed text and handwritten text) and cross modal results demonstrate that this is feasible, and open different research lines. Furthermore, two architectures for selective style transfer, which means transferring style to only desired image pixels, are proposed. Finally, scene text selective style transfer is evaluated as a data augmentation technique to expand scene text detection datasets, resulting in a boost of text detectors performance. Our implementation of the described models is publicly available.