Exploring Font-independent Features for Scene Text Recognition
arXiv:2009.07447
Abstract
Scene text recognition (STR) has been extensively studied in last few years. Many recently-proposed methods are specially designed to accommodate the arbitrary shape, layout and orientation of scene texts, but ignoring that various font (or writing) styles also pose severe challenges to STR. These methods, where font features and content features of characters are tangled, perform poorly in text recognition on scene images with texts in novel font styles. To address this problem, we explore font-independent features of scene texts via attentional generation of glyphs in a large number of font styles. Specifically, we introduce trainable font embeddings to shape the font styles of generated glyphs, with the image feature of scene text only representing its essential patterns. The generation process is directed by the spatial attention mechanism, which effectively copes with irregular texts and generates higher-quality glyphs than existing image-to-image translation methods. Experiments conducted on several STR benchmarks demonstrate the superiority of our method compared to the state of the art.
arXiv commentaccepted by ACM MM 2020. Code: https://actasidiot.github.io/EFIFSTR/
References in corpus (8)
- Conditional Generative Adversarial Nets
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- Focusing Attention: Towards Accurate Text Recognition in Natural Images
- COCO-Text: Dataset and Benchmark for Text Detection and Recognition in Natural Images
- Synthetic Data for Text Localisation in Natural Images
- Recursive Recurrent Nets with Attention Modeling for OCR in the Wild
- Edit Probability for Scene Text Recognition
- NRTR: A No-Recurrence Sequence-to-Sequence Model For Scene Text Recognition