IT-710-Project-Video-QnA-System

About:

• The aim of our project is that we want to perform a semantic search on videos in a computationally efficient manner with the help of the URL provided by the user and to solve all the problems identified by making a simple tool. This project focuses on new techniques and approaches to develop a video question answering system.

• You can find an introductory video up here.

• You can find the project report up here.

How to setup:

Currently we only support Linux and MacOS.

# Clone our repository
$ git clone https://github.com/programmer290399/IT-710-Project-Video-QnA-System.git

# Navigate to the project root
$ cd IT-710-Project-Video-QnA-System

# Create a virtual env
$ python3 -m venv .venv

# Activate the virtual env
$ . .venv/bin/activate 

# Install the dependencies
$ pip install -r requirements.txt

# Run the server
$ python manage.py runserver

Citation:

@article{Seo2017BidirectionalAF,
    author = {Minjoon Seo and Aniruddha Kembhavi and Ali Farhadi and Hannaneh Hajishirzi},
    journal = {ArXiv},
    title = {Bidirectional Attention Flow for Machine Comprehension},
    volume = {abs/1611.01603},
    year = {2017}
}