The Omniglot Challenge (paper)
Own Model for the Omniglot Challenge.
Brain & Cognitive Society, IIT Kanpur
- Convert all stroke data to 25-point splines. ✅
- Generate a b-vector using a variational autoencoder for each and every such stroke. ✅
- Use clustering to get number of primitives, and then turn them into vectors for each and every image in the background set using one-hot encoding. ✅
- Perform supervised learning using a Convolutional Neural Network on the images and respective vectors to get a network which maps your character images to stroke data latent vector space.(In Progress)
- Use the trained model for classification or one-shot learning tasks by introducing another model which maps the latent space of the image vector to desired output.
- Report acquired results.