google-research-datasets / vrdu
We identify the desiderata for a comprehensive benchmark and propose Visually Rich Document Understanding (VRDU). VRDU contains two datasets that represent several challenges: rich schema including diverse data types, complex templates, and diversity of layouts within a single document type.
☆74Updated last year
Related projects ⓘ
Alternatives and complementary repositories for vrdu
- ReadingBank: A Benchmark Dataset for Reading Order Detection☆91Updated 2 months ago
- The code related to the baselines from NeurIPS 2021 paper "DUE: End-to-End Document Understanding Benchmark."☆35Updated last year
- ☆30Updated 7 months ago
- Algorithms, papers, datasets, performance comparisons for Document AI. Continuously updating.☆165Updated this week
- Dataset of PNG images from synthetically generated table layouts with annotations in JSONL files☆129Updated last year
- Publicly released code for the LAMBERT model☆102Updated 3 years ago
- DocILE: Document Information Localization and Extraction Benchmark☆118Updated 6 months ago
- InstructDoc: A Dataset for Zero-Shot Generalization of Visual Document Understanding with Instructions (AAAI2024)☆146Updated 5 months ago
- ☆55Updated 3 years ago
- Example codebase for fine-tuning layoutLMv3 on DocVQA☆49Updated 2 years ago
- Doc2Graph transforms documents into graphs and exploit a GNN to solve several tasks.☆116Updated last year
- High-Performance Transformers for Table Structure Recognition Need Early Convolutions☆41Updated 7 months ago
- CTE: Contextualized Table Extraction Dataset☆17Updated last year
- ☆74Updated 2 years ago
- My implementation of Kosmos2.5 from the paper: "KOSMOS-2.5: A Multimodal Literate Model"☆68Updated last week
- Dataset and scripts for HRDoc☆33Updated last year
- ☆155Updated last year
- CVPR 2022: Table Structure Recognition☆39Updated 2 years ago
- ☆54Updated 10 months ago
- Official implementation for Dessurt☆56Updated last year
- An unofficial Implementation of DocParser: End-to-end OCR-free Information Extraction from Visually Rich Documents☆33Updated last year
- an unofficial code for augment-XY-CUT in XYLayoutLM☆25Updated 2 years ago
- An unofficial Pytorch implementation of ERNIE-Layout which is originally released through PaddleNLP.☆99Updated last year
- OCR Annotations from Amazon Textract for Industry Documents Library☆99Updated 2 years ago
- ☆41Updated 2 years ago
- The WordScape repository contains code for the WordScape pipeline to create datasets to train document understanding models.☆32Updated 11 months ago
- Visually-Situated Natural Language Understanding with Contrastive Reading Model and Frozen Large Language Models, EMNLP 2023☆43Updated 5 months ago
- ☆242Updated last year
- XFUND: A Multilingual Form Understanding Benchmark☆186Updated 2 years ago
- Incorporating VIsual LAyout Structures for Scientific Text Classification☆173Updated last year