PingAn-VCGroup's Solution for ICDAR 2021 Competition on Scientific Literature Parsing Task B: Table Recognition to HTML
arXiv:2105.01848
Abstract
This paper presents our solution for ICDAR 2021 competition on scientific literature parsing taskB: table recognition to HTML. In our method, we divide the table content recognition task into foursub-tasks: table structure recognition, text line detection, text line recognition, and box assignment.Our table structure recognition algorithm is customized based on MASTER [1], a robust image textrecognition algorithm. PSENet [2] is used to detect each text line in the table image. For text linerecognition, our model is also built on MASTER. Finally, in the box assignment phase, we associatedthe text boxes detected by PSENet with the structure item reconstructed by table structure prediction,and fill the recognized content of the text line into the corresponding item. Our proposed methodachieves a 96.84% TEDS score on 9,115 validation samples in the development phase, and a 96.32%TEDS score on 9,064 samples in the final evaluation phase.
arXiv comment8 Pages, 7 Figures
References in corpus (6)
- Lookahead Optimizer: k steps forward, 1 step back
- MASTER: Multi-Aspect Non-local Network for Scene Text Recognition
- Image-based table recognition: data, model, and evaluation
- Exploiting Deep Representations for Neural Machine Translation
- Learning Graph Normalization for Graph Neural Networks
- PingAn-VCGroup's Solution for ICDAR 2021 Competition on Scientific Table Image Recognition to Latex
Cited by in corpus (7)
- A large-scale dataset for end-to-end table recognition in the wild
- An End-to-End Multi-Task Learning Model for Image-based Table Recognition
- Rethinking Image-based Table Recognition Using Weakly Supervised Methods
- PingAn-VCGroup's Solution for ICDAR 2021 Competition on Scientific Table Image Recognition to Latex
- SPRINT: Script-agnostic Structure Recognition in Tables
- 1st Place Solution for ICDAR 2021 Competition on Mathematical Formula Detection
- Synthesizing Realistic Data for Table Recognition