Video Gesture Analysis for Autism Spectrum Disorder Detection

Reference

Code for the paper Video Gesture Analysis for Autism Spectrum Disorder Detection, Zunino et al, ICPR 2018 (pdf, proceedings)

@inproceedings{2018_ICPR,
author = {Andrea Zunino and Pietro Morerio and Andrea Cavallo and Caterina Ansuini and Jessica Podda and Francesca Battaglia and Edvige Veneselli and Cristina Becchio and Vittorio Murino},
title = {Video Gesture Analysis for Autism Spectrum Disorder Detection},
booktitle = {International Conference on Pattern Recognition (ICPR)},
year = {2018},
}

The dataset

The dataset contains trials performed by both autistic kids and control subjects. Main task is to perform this two-class classification.

The code

This repo is based on this public code implementing an attention-based LSTM and is meant to classify autistic kids from the kinematics of motion. It contains:

• matlab scripts for estracting CNN features from the video frames (GoogleNet - Caffe)
• python scripts for the actual LSTM training (theano)

(Main) Dependencies

• Python 2.7
• NumPy
• scikit learn
• skimage
• Theano
• h5py
• Matlab
• Caffe

License

This repository is released under the MIT LICENSE.