TANDEM (0.5)

Let’s build a GUI that combines the power of Google’s Tesseract OCR and FeatureExtractor.

The idea is to build an environment (web-based or standalone) where you can take your text overlayed object, toss it in, and have a save ready output file to take away with you. Generate the data you need to visualize, explore, parse apart, and build the story. There is an interesting dialogue between text and images happening in comics, children’s picture books, marketing materials, illustrated maps, illuminated manuscripts, etc. Get your data, understand the output variables in simple and easy to reference ways, and get back to finding your story.

  • Team Note: Be ready to learn. Everyone involved in this intimidating project will bleed through their role and engage in collaborative learning of each of the elements needed to complete the project. Developers must understand design, designers must understand branding, outreach/branding must understand how the thing works, and the project coordinator must understand how to get conversations rolling to hit deadlines.
  • Developer:  Bravery. Develop clean and clear code that will allow us to wrap our OCR and Image Processing Software as modules to be placed in the overall software. Working understanding of code and willingness to dedicate time to digging into what needs to be written to get this off the ground. Knowledge of Python or a single language at the very least.
  • Designer: Understand user interaction and develop aesthetically simple, intuitive interface. Understand design basics, have a working proficiency in Adobe Design programs. Also maintain brand identity in conjunction with Outreach Coordinator.
  • Outreach Coordinator: Social butterflies. We need community support. Work on creating a voice and an audience for this project. Using not only social media but having the ability to track where our message is working best. What tweets work, what outlets are giving good feedback. We need to make a communal conversation that helps us reach our goal.
  • Project Manager: Keep your hand on the pulse of the schedule, set deadlines, gather learning resources, keep open lines of communication between team members. I have so many people in mind for this and each one can potentially bring an entirely different outcome to the project. I want a project manager who wants to see this thing materialize.

Reassurance – Let’s say we build it in Python and Javascript – Here are key some pieces we can consider:

Google’s Tesseract OCR

Python-tesseract is a python wrapper for google’s Tesseract-OCR

FeatureExtractor (Let’s talk to Lev about this. It is one of his tools afterall.) –

PyJamas GUI Toolkit

4 thoughts on “TANDEM (0.5)

  1. Pingback: Entry 1 | DH Praxis 15 - Chris Vitale

Comments are closed.