FUNSD: A Dataset for Form Understanding in Noisy Scanned Documents

Jaume, Guillaume; Ekenel, Hazım Kemal; Thiran, Jean‐Philippe

doi:10.1109/icdarw.2019.10029

Cited by 210 publications

(178 citation statements)

References 16 publications

Supporting

Mentioning

178

Contrasting

Order By: Relevance

“…We validate our method, DocStruct, on two benchmarks, MedForm and FUNSD (Jaume et al, 2019). The first one is built by us and composed of medical examination reports, and the second is composed of various real, fully annotated, scanned forms.…”

Section: Check Items Resultsmentioning

confidence: 99%

See 1 more Smart Citation

DocStruct: A Multimodal Method to Extract Hierarchy Structure in Document for General Form Understanding

Wang¹,

Zhan²,

Liu³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

Form understanding depends on both textual contents and organizational structure. Although modern OCR performs well, it is still challenging to realize general form understanding because forms are commonly used and of various formats. The table detection and handcrafted features in previous works cannot apply to all forms because of their requirements on formats. Therefore, we concentrate on the most elementary components, the key-value pairs, and adopt multimodal methods to extract features. We consider the form structure as a tree-like or graph-like hierarchy of text fragments. The parent-child relation corresponds to the key-value pairs in forms. We utilize the state-of-the-art models and design targeted extraction modules to extract multimodal features from semantic contents, layout information, and visual images. A hybrid fusion method of concatenation and feature shifting is designed to fuse the heterogeneous features and provide an informative joint representation. We adopt an asymmetric algorithm and negative sampling in our model as well. We validate our method on two benchmarks, MedForm and FUNSD, and extensive experiments demonstrate the effectiveness of our method.

show abstract

Section: Check Items Resultsmentioning

confidence: 99%

“…• FUNSD-base: The baseline is offered in Jaume et al (2019), which uses semantic and layout information.…”

Section: Ablation Study and Baseline Comparisonmentioning

confidence: 99%

DocStruct: A Multimodal Method to Extract Hierarchy Structure in Document for General Form Understanding

Wang¹,

Zhan²,

Liu³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

show abstract

“…We evaluate how reading order impacts overall performance of graph-based information extraction from form-like documents. We adopt two form understanding tasks as Jaume et al (2019), including word labeling and word grouping. Word labeling is the task of assigning each word a label from a set of predefined entity categories, realized by node classification.…”

Section: Methodsmentioning

confidence: 99%

“…FUNSD. FUNSD (Jaume et al, 2019) is a public dataset for form understanding in noisy scanned documents, containing a collection of research, marketing, and advertising documents that vary widely in their structure and appearance. The dataset consists of 199 annotated forms with 9,707 entities and 31,485 word-level annotations for 4 entity types: header, question, answer, and other.…”

Section: Datasetsmentioning

confidence: 99%

ROPE: Reading Order Equivariant Positional Encoding for Graph-based Document Information Extraction

Lee¹,

Li²,

Wang³

et al. 2021

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer

View full text Add to dashboard Cite

Natural reading orders of words are crucial for information extraction from form-like documents. Despite recent advances in Graph Convolutional Networks (GCNs) on modeling spatial layout patterns of documents, they have limited ability to capture reading orders of given word-level node representations in a graph. We propose Reading Order Equivariant Positional Encoding (ROPE), a new positional encoding technique designed to apprehend the sequential presentation of words in documents. ROPE generates unique reading order codes for neighboring words relative to the target word given a word-level graph connectivity. We study two fundamental document entity extraction tasks including word labeling and word grouping on the public FUNSD dataset and a large-scale payment dataset. We show that ROPE consistently improves existing GCNs with a margin up to 8.4% F1-score. * Work done while an intern at Google Research.

show abstract

“…The study [7] presented a Form Understanding (FUNSD) dataset containing 199 completely annotated noisy scanned forms in JSON format. FUNSD consists of forms with diverse fields like Marketing, Advertisement, Science, and few others, which are used for text detection, Optical Character Recognition (OCR), and document layout understanding tasks.…”

Section: A Standard Datasetsmentioning

confidence: 99%

Multi-Layout Unstructured Invoice Documents Dataset: A Dataset for Template-Free Invoice Processing and Its Evaluation Using AI Approaches

2021

View full text Add to dashboard Cite

The daily transaction of an organization generates a vast amount of unstructured data such as invoices and purchase orders. Managing and analyzing unstructured data is a costly affair for the organization. Unstructured data has a wealth of hidden valuable information. Extracting such insights automatically from unstructured documents can significantly increase the productivity of an organization. Thus, there is a huge demand to develop a tool that can automate the extraction of key fields from unstructured documents. Researchers have used different approaches for extracting key fields, but the lack of annotated and highquality datasets is the biggest challenge. Existing work in this area has used standard and custom datasets for extracting key fields from unstructured documents. Still, the existing datasets face some serious challenges, such as poor-quality images, domain-related datasets, and a lack of data validation approaches to evaluate data quality. This work highlights the detailed process flow for endto-end key fields extraction from unstructured documents. This work presents a high-quality, multi-layout unstructured invoice documents dataset assessed with a statistical data validation technique. The proposed multi-layout unstructured invoice documents dataset is highly diverse in invoice layouts to generalize key field extraction tasks for unstructured documents. The proposed multilayout unstructured invoice documents dataset is evaluated with various feature extraction techniques such as Glove, Word2Vec, FastText, and AI approaches such as BiLSTM and BiLSTM-CRF. We also present the comparative analysis of feature extraction techniques and AI approaches on the proposed multi-layout unstructured invoice document dataset. We attained the best results with BiLSTM-CRF model. INDEX TERMS Artificial Intelligence (AI), information extraction, key field extraction, Named Entity Recognition (NER), template-free invoice processing, unstructured data.

show abstract

FUNSD: A Dataset for Form Understanding in Noisy Scanned Documents

Cited by 210 publications

References 16 publications

DocStruct: A Multimodal Method to Extract Hierarchy Structure in Document for General Form Understanding

DocStruct: A Multimodal Method to Extract Hierarchy Structure in Document for General Form Understanding

ROPE: Reading Order Equivariant Positional Encoding for Graph-based Document Information Extraction

Multi-Layout Unstructured Invoice Documents Dataset: A Dataset for Template-Free Invoice Processing and Its Evaluation Using AI Approaches

Contact Info

Product

Resources

About