GridFormer: Towards Accurate Table Structure Recognition via Grid Prediction

TL;DR

GridFormer leverages grid prediction with a DETR-style recognizer to accurately interpret complex table structures, achieving competitive results across diverse challenging benchmarks.

Contribution

The paper introduces a flexible grid-based table representation and a single-shot DETR-style recognizer for efficient, accurate table structure recognition.

Findings

Outperforms existing methods on five challenging benchmarks.

Effectively handles wired, wireless, multi-merge-cell, oriented, and distorted tables.

Demonstrates robustness and versatility in table structure recognition.

Abstract

All tables can be represented as grids. Based on this observation, we propose GridFormer, a novel approach for interpreting unconstrained table structures by predicting the vertex and edge of a grid. First, we propose a flexible table representation in the form of an MXN grid. In this representation, the vertexes and edges of the grid store the localization and adjacency information of the table. Then, we introduce a DETR-style table structure recognizer to efficiently predict this multi-objective information of the grid in a single shot. Specifically, given a set of learned row and column queries, the recognizer directly outputs the vertexes and edges information of the corresponding rows and columns. Extensive experiments on five challenging benchmarks which include wired, wireless, multi-merge-cell, oriented, and distorted tables demonstrate the competitive performance of our model over other methods.