Robust Table Structure Recognition with Dynamic Queries Enhanced Detection Transformer
Jiawei Wang, Weihong Lin, Chixiang Ma, Mingze Li, Zheng Sun, Lei Sun,, Qiang Huo

TL;DR
This paper introduces TSRFormer, a novel table structure recognition method that uses a dynamic queries enhanced DETR framework for direct separation line prediction, achieving state-of-the-art results on multiple benchmarks.
Contribution
The paper proposes a new line regression formulation for TSR, with innovative dynamic query design and progressive regression techniques to improve accuracy and efficiency.
Findings
Achieves state-of-the-art performance on benchmark datasets.
Demonstrates robustness to complex, distorted, and curved table structures.
Improves localization accuracy over previous methods.
Abstract
We present a new table structure recognition (TSR) approach, called TSRFormer, to robustly recognizing the structures of complex tables with geometrical distortions from various table images. Unlike previous methods, we formulate table separation line prediction as a line regression problem instead of an image segmentation problem and propose a new two-stage dynamic queries enhanced DETR based separation line regression approach, named DQ-DETR, to predict separation lines from table images directly. Compared to Vallina DETR, we propose three improvements in DQ-DETR to make the two-stage DETR framework work efficiently and effectively for the separation line prediction task: 1) A new query design, named Dynamic Query, to decouple single line query into separable point queries which could intuitively improve the localization accuracy for regression tasks; 2) A dynamic queries based…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsHandwritten Text Recognition Techniques · Advanced Image and Video Retrieval Techniques · Image Retrieval and Classification Techniques
MethodsMulti-Head Attention · Attention Is All You Need · Linear Layer · Label Smoothing · Residual Connection · Byte Pair Encoding · Dropout · Layer Normalization · Dense Connections · Position-Wise Feed-Forward Layer
