Hemank Lamba, Shashank Srikanth *, Dheeraj Reddy *, Shwetanshu Singh, Karandeep Juneja, Ponnurangam Kumaraguru
Drinking & Driving | Passenger Distraction | Distracted Driving on Bikes |
---|---|---|
![]() |
![]() |
![]() |
This repository contains code and data required to reproduce the results of "Driving the Last Mile: Characterizing and Understanding Distracted Driving Posts on Social Networks". (Paper link)
Our work deals with identifying distracted driving content on social media (Snapchat) using computer vision methods and analyzing the data for spatio-temporal patterns and charecterizing the users demographics.
A few examples of distracted driving content on Snapchat are given in the figure above.
The code has been tested with python3
and PyTorch 0.4.1
. The codebase can also support PyTorch 1.0
with slight modifications.
virtualenv -p python3.5 venv
source venv/bin/activate
pip install -r requirements.txt
The codebase is divided into the following major portions:
All the pretrained models and the Snapchat scraper code are available on request via email at pk [at] iiitd [dot] ac [dot] in
.
For additional details, plots, and discussions, refer to the project page here.
If you are using this code or the pre-trained models, please cite our paper as follows:
@article{lambadistracteddriving,
author={Hemank Lamba, Shashank Srikanth, Dheeraj Reddy Pailla, Shwetanshu Singh, Karandeep Juneja, and Ponnurangam Kumaraguru},
title={Driving the Last Mile: Characterizing and Understanding Distracted Driving Posts on Social Networks},
booktitle={Proceedings of the International AAAI Conference on Web and Social Media (ICWSM) 2020},
year={2020},
}