ronojoy/indus-script-ocr

The Indus script optical grapheme recognition engine (from archaeological artifact images)

Indus Script OCR

To automatically locate text patches/regions, segment individual symbols/characters from those regions and also identify each symbol/character belonging to the Indus Script, given images of Indus seals from archaeological sites, using image processing and deep learning techniques. [WIP]

View our research article titled "Deep Learning the Indus Script" arXived at: arXiv:1702.00523v1

Deploying the app

• Setup the GPU machine to run the service,

  • Install latest nvidia drivers, from http://www.geforce.com/drivers
  • Install the nvidia-docker plug-in over docker, from https://github.com/NVIDIA/nvidia-docker/releases
  • Make sure you have git-lfs installed (https://git-lfs.github.com/)

• Launch the service,

  • Build the docker image: nvidia-docker build --no-cache=true -t indus-script-ocr:latest .
  • To launch a docker container: nvidia-docker run -it -v "$PWD":/root/workspace --rm --env-file app.env --name indus-script-ocr-service indus-script-ocr:latest

Press Coverage:

• The Verge
• The Hindu
• Times of India
• SBS Radio, Australia

Talks

• Indian Deep Learning Initiative (IDLI): slide deck, video, link
• ThoughtWorks Geek Night: slide deck, video, link
• ChennaiPy: link
• Anthill Inside 2017: proposal

Citation

Please cite indus-script-ocr in your publications if it helps your research:

@article{palaniappan2017deep,
title={Deep Learning the Indus Script},
author={Palaniappan, Satish and Adhikari, Ronojoy},
journal={arXiv preprint arXiv:1702.00523},
year={2017}
}