Awesome Table Structure Recognition

Datasets

1. ICDAR-2013 (Website, Paper)
• # Test = 156
• Table detection, Table Recognition
• 정부 문서 구글링
데이터셋 예시


 
 

2. ICDAR 2019 cTDaR (Website, Paper, Github)
• ICDAR 2019 Competition on Table Detection and Recognition (cTDaR)
• 고문서 1,000장 / 현대 문서 1,000장 각각 존재
• Table detection, Table Recognition
데이터셋 예시





 

3. SciTSR (Github, Paper)
• # Train / # Test = 12K / 3K
• Complicated # Train / # Test = 2,885 / 716
• 아카이브의 학술논문 테이블
데이터셋 예시









 

4. TableBank (Website, Github, Paper)
• # Train / val / test = 130K / 10K / 5K (Table Structure Recognition)
• 인터넷에서 .docx 파일 크롤링(다국어) + arxiv.org 에서 .tex 파일 다운로드
• Task: 이미지 to HTML
 

5. PubTabNet (Website, Github, Paper)
• 568K table images
• GT format: HTML
• Source: PubMed Central Open Access Subset (PMCOA)
데이터셋 예시
좌: 이미지 / 우: HTML 렌더링

 
 

6. FinTabNet (Website, Paper)
• Train - 92K / Test - 11K / Val - 11K
• Cell structure 인식이 주요 목표임.
• S&P500 회사들의 annual report 에서 테이블 추출
• 학술 논문에 비해 난이도가 더 높은 것 같음
데이터셋 예시





 

7. PubTables-1M (Github, Paper)
• ~1M tables, 460K pages
• Task 1: Table Detection (TD)
• Task 2: Table Structure Recognition (TSR)
• Task 3: Funcional Analysis (FA)
→ Header, Cell 등을 정의



• Object detection 모델을 사용한 cell detection으로 PubTables-1M 데이터셋에 실험

 

8. WTW (Github, Paper)
• 14K tables
• 상품 등 natural scene 의 table 이미지 위주. 인쇄된 문서 포함.
• Cycle-CenterNet 제안
데이터셋 예시



 
 

9. ICDAR 2021 (Paper, )
• ICDAR 2021 Competition on Scientific Table Image Recognition to LaTeX
• 표 이미지를 LATEX 코드로 변환하는 Task.
• Task 1: Table Structure Reconstruction (TSR)
• Task 2: Table Content Reconstruction (TCR)
• 데이터셋 규모

데이터셋 예시



 

10. TabLeX (Google Drive, Paper)
• 1M+ tables
• ICDAR 2021과 유사 (table structure, table content 분리)
• Table image → LaTeX
데이터셋 예시

 

Metric

1. TEDS
Tree-edit-distance-based similarity
Paper: Image-based table recognition: data, model, and evaluation



표를 표현하는 HTML 코드를 tree 구조로 만들고, 이 트리들 간의 Tree Edit Distance(Paper)를 계산.
Edit distance를 정규화하여 TEDS 값 산출.

2. BLEU score
https://en.wikipedia.org/wiki/BLEU
 

Papers

• (PubTabNet, EDD, TEDS score) Image-based table recognition: data, model, and evaluation (ECCV 2020)

• (TabStruct-Net) Table Structure Recognition using Top-Down and Bottom-Up Cues (ECCV 2020)

• (FinTabNet, GTE) Global Table Extractor (GTE): A Framework for Joint Table Identification and Cell Structure Recognition Using Visual Context (WACV 2021)

• (DGCNN) Rethinking Table Recognition using Graph Neural Networks (ICDAR 2019) - Ref. 50

• (GraphTSR, SciTSR) Complicated Table Structure Recognition

• Parsing Table Structures in the Wild (ICCV 2021)

• TGRNet: A Table Graph Reconstruction Network for Table Structure Recognition. (ICCV 2021)

• Visual Understanding of Complex Table Structures from Document Images (WACV 2022)

• Flexible Table Recognition and Semantic Interpretation System (arXiv:2105.11879)

• (NCGM) Neural Collaborative Graph Machines for Table Structure Recognition (arXiv:2111.13359)

• LGPMA: Complicated Table Structure Recognition with Local and Global Pyramid Mask Alignment (ICDAR 2021)

Resources

1. IBM - Tutorial on Table Extraction and Understanding for Scientific and Enterprise Applications

4. https://github.com/doc-analysis

5. https://www.microsoft.com/en-us/research/project/document-ai/
https://arxiv.org/pdf/2111.08609.pdf