Dataset Overview

Dataset Description

• Source: This dataset is extracted from the InfoVQA dataset test split, containing infographics collected from the internet by searching for “infographics”. Questions and answers are manually annotated.

Data Structure

• Features:
  • questionId: Question ID (string)
  • query: Query content (string)
  • answer: Answer (empty)
  • answer_type: Answer type (empty)
  • image: Image (image)
  • image_filename: Image filename (string)
  • operation/reasoning: Operation/Reasoning (empty)
  • ocr: OCR text (string)
  • data_split: Data split (string)
  • source: Data source (string)

Data Split

• Test set:
  • test: Contains 500 samples, total size 277,995,931 bytes.

Dataset Size

• Download size: 218,577,138 bytes.
• Dataset size: 277,995,931 bytes.

Data Loading

• Loading method:

from datasets import load_dataset

ds = load_dataset("vidore/infovqa_test_subsampled", split="test")

Citation Information

• Citation format:

@misc{mathew_infographicvqa_2021,
  title = {{InfographicVQA}},
  copyright = {arXiv.org perpetual, non-exclusive license},
  url = {https://arxiv.org/abs/2104.12756},
  doi = {10.48550/ARXIV.2104.12756},
  urldate = {2024-06-02},
  publisher = {arXiv},
  author = {Mathew, Minesh and Bagal, Viraj and Tito, Rubèn Pérez and Karatzas, Dimosthenis and Valveny, Ernest and Jawahar, C. V},
  year = {2021},
  note = {Version Number: 2},
  keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, Computer Vision and Pattern Recognition (cs.CV)},
}