In this paper, we present StrucTexTv2, an effective document image pre-training framework, by performing masked visual-textual prediction. It consists of two self-supervised pre-training tasks: masked image modeling and masked language modeling, based on text region-level image masking. The proposed method randomly masks some image regions according to the bounding box coordinates of text words. The objectives of our pre-training tasks are reconstructing the pixels of masked image regions and the corresponding masked tokens simultaneously. Hence the pre-trained encoder can capture more textual semantics in comparison to the masked image modeling that usually predicts the masked image patches. Compared to the masked multi-modal modeling methods for document image understanding that rely on both the image and text modalities, StrucTexTv2 models image-only input and potentially deals with more application scenarios free from OCR pre-processing. Extensive experiments on mainstream benchmarks of document image understanding demonstrate the effectiveness of StrucTexTv2. It achieves competitive or even new state-of-the-art performance in various downstream tasks such as image classification, layout analysis, table structure recognition, document OCR, and information extraction under the end-to-end scenario.
@article{arxiv.2303.00289,
title = {StrucTexTv2: Masked Visual-Textual Prediction for Document Image Pre-training},
author = {Yuechen Yu and Yulin Li and Chengquan Zhang and Xiaoqiang Zhang and Zengyuan Guo and Xiameng Qin and Kun Yao and Junyu Han and Errui Ding and Jingdong Wang},
journal= {arXiv preprint arXiv:2303.00289},
year = {2023}
}