Skip to content

Latest commit

 

History

History
40 lines (34 loc) · 668 Bytes

README.md

File metadata and controls

40 lines (34 loc) · 668 Bytes
Raw

Interactive Doodle Recognition

This work is a attempt to train a Deep Neural Network to classify doodles from the Google's QuickDraw database as the user interacts with the application.

Available predictions

  • Angel;
  • Bicycle;
  • Cat;
  • Clock;
  • Cup;
  • Diamond;
  • Dolphin;
  • Door;
  • Eye;
  • Eyeglasses;
  • Hat;
  • Headphones;
  • Helicopter;
  • House;
  • Octopus;
  • Plant;
  • Popsicle;
  • Ship;
  • Umbrella;
  • Windmill;

Dependencies:

  • raysan5/raylib;
  • libeigen/eigen;
  • opencv/opencv;
  • Dobiasd/FunctionalPlus
  • Dobiasd/frugally-deep;
  • nlohmann/json;

TODO

  • For now, there are only 20 possible predictions;
  • Make a better brush.