Chargrid: Towards Understanding 2D Documents
arXiv:1809.08799
Abstract
We introduce a novel type of text representation that preserves the 2D layout of a document. This is achieved by encoding each document page as a two-dimensional grid of characters. Based on this representation, we present a generic document understanding pipeline for structured documents. This pipeline makes use of a fully convolutional encoder-decoder network that predicts a segmentation mask and bounding boxes. We demonstrate its capabilities on an information extraction task from invoices and show that it significantly outperforms approaches based on sequential text or document images.
arXiv commentTo be published at EMNLP 2018
References in corpus (6)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Focal Loss for Dense Object Detection
- ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
- Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks
- Learning to Extract Semantic Structure from Documents Using Multimodal Fully Convolutional Neural Network
- Multilingual Language Processing From Bytes
Cited by in corpus (8)
- Kleister: A novel task for Information Extraction involving Long Documents with Complex Layout
- MMOCR: A Comprehensive Toolbox for Text Detection, Recognition and Understanding
- PICK: Processing Key Information Extraction from Documents using Improved Graph Learning-Convolutional Networks
- TRIE: End-to-End Text Reading and Information Extraction for Document Understanding
- StructuralLM: Structural Pre-training for Form Understanding
- Robust Layout-aware IE for Visually Rich Documents with Pre-trained Language Models
- Key Information Extraction From Documents: Evaluation And Generator
- Graph Convolution for Multimodal Information Extraction from Visually Rich Documents