Skip to content

Cuneiform character recognition using shape contexts.

Notifications You must be signed in to change notification settings

edwardclem/santak

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

santak

santak; noun. "cuneiform wedge", Foxvog 2016.

Drawing-based lookup app for cuneiform characters. Uses shape context matching to compute distance between drawn images and reference characters. Reference characters from CuneiformComposite font, provided by the Unicode consortium.

Inspired by ShapeCatcher.

setup/installation

All dependencies can be handled by the Conda package manager. Install Conda, then run: conda env create -f environment.yml

To use the prototype generation workflow below, run pip install -e . from within the directory.

Prototype generation

Prototype images for shape matching are created using the Luigi workflow found in santak/datagen that renders characters in the Cuneiform unicode code block and produces contours using a Canny edge filter. See scripts/aggregate.sh for a usage example.

Usage

Run ./santakgui -p <prototyes file>

TODO

- UI Improvements
- Speed/parallelization

About

Cuneiform character recognition using shape contexts.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published