MaskOCR: Scene Text Recognition with Masked Vision-Language Pre-training

Text images contain both visual and linguistic information. However, existing pre-training techniques for text recognition mainly focus on either visual representation learning or linguistic knowledge learning. In this paper, we propose a novel approach to unify vision and language pre-training in the classical encoder-decoder recognition framework. We adopt the masked image modeling approach to pre-train the feature encoder using a large set of unlabeled real text images, which allows us to learn strong visual representations. In contrast to introducing linguistic knowledge with an additional language model, we directly pre-train the sequence decoder. Specifically, we transform text data into synthesized text images to unify the data modalities of vision and language, and enhance the language modeling capability of the sequence decoder using a proposed masked image-language modeling scheme. Significantly, the encoder is frozen during the pre-training phase of the sequence decoder. Experimental results demonstrate that our proposed method achieves superior performance on benchmark datasets, including Chinese and English text images. The code for our approach will be made available.

Authors

• Pengyuan Lyu
• Chengquan Zhang
• Shanshan Liu
• Meina Qiao
• Yangliu Xu
• Liang Wu Southwestern University of Finance and Economics Big Data Laboratory on Financial Security and Behavior, SWUFE
• Kun Yao Baidu
• Junyu Han Chongqing Changan Automobile Co., Ltd.
• Errui Ding Baidu
• Jingdong Wang Baidu

Keywords

No keywords are indexed for this paper.