metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype: string
  splits:
    - name: train
      num_bytes: 12173747703
      num_examples: 7224600
    - name: val
      num_bytes: 1352108669.283
      num_examples: 802733
    - name: test
      num_bytes: 1484450563.896
      num_examples: 891924
  download_size: 12115256620
  dataset_size: 15010306936.179
task_categories:
  - image-to-text
language:
  - en
size_categories:
  - 1M<n<10M
pretty_name: MJSynth

Dataset Card for "MJSynth_text_recognition"

This is the MJSynth dataset for text recognition on document images, synthetically generated, covering 90K English words. It includes training, validation and test splits. Source of the dataset: https://www.robots.ox.ac.uk/~vgg/data/text/

Use dataset streaming functionality to try out the dataset quickly without downloading the entire dataset (refer: https://huggingface.co/docs/datasets/stream)

Citation details provided on the source website (if you use the data please cite):

@InProceedings{Jaderberg14c, author = "Max Jaderberg and Karen Simonyan and Andrea Vedaldi and Andrew Zisserman", title = "Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition", booktitle = "Workshop on Deep Learning, NIPS", year = "2014", }

@Article{Jaderberg16, author = "Max Jaderberg and Karen Simonyan and Andrea Vedaldi and Andrew Zisserman", title = "Reading Text in the Wild with Convolutional Neural Networks", journal = "International Journal of Computer Vision", number = "1", volume = "116", pages = "1--20", month = "jan", year = "2016", }