In this project, we are going to implement a neural network for recognizing Grammatical Facial Expressions (GFEs) used in the Brazilian Sign Language. To do so we'll use ConvNetJS, a Javascript library, to train a neural network using backpropagation algorithms.
- Grammatical Facial Expressions Data Set
- Relevant Paper
- ConvNetJS - Javascript library for training Neural Networks
- Notes of the Stanford CS class CS231n: Convolutional Neural Networks for Visual Recognition
Contains a few examples of using Convuntional Neural Networks. ConvNetJS implements Deep Learning models and learning algorithms as well as nice browser-based demos, all in Javascript.
Pre-processed data in json format.
Grammatical Facial Expressions for Brazilian Sign Language
The dataset is organized in 36 files: 18 datapoint files and 18 target files, one pair for each video which compose the dataset.The name of the file refers to each video: the letter corresponding to the user (A and B), name of grammatical facial expression and a specification (target or datapoints).
Contains:
-
Datapoints files (* _ datapoints.txt): a timestamp (double) and 100 numeric attributes (double)
-
Targets files (* _ targets.txt): a class attribute (interger)
Small CLI style node app to convert delimited .txt files into .json
Project reports.
Our implementation of a neural network using ConvNetJS to rate facial expressions.