Real-World Font Recognition Using Deep Network and Domain Adaptation
This work solves a fine-grained classification challenge for applications in document analysis and design, but it is incremental as it builds on existing domain adaptation methods.
The paper tackles the problem of font recognition from real-world text images by addressing the domain gap between synthetic and real data, achieving over 80% top-5 accuracy on a real-world dataset.
We address a challenging fine-grain classification problem: recognizing a font style from an image of text. In this task, it is very easy to generate lots of rendered font examples but very hard to obtain real-world labeled images. This real-to-synthetic domain gap caused poor generalization to new real data in previous methods (Chen et al. (2014)). In this paper, we refer to Convolutional Neural Networks, and use an adaptation technique based on a Stacked Convolutional Auto-Encoder that exploits unlabeled real-world images combined with synthetic data. The proposed method achieves an accuracy of higher than 80% (top-5) on a real-world dataset.