An innovative solution to the issue of remote exam proctoring is the development of AI-based monitoring systems that utilize machine learning algorithms. These systems have the ability to automatically identify and flag potential cheating behaviors, such as irregular eye movements, head gestures, and keyboard activity, and alert the proctor in real-time. To this end, we plan to create an AI monitoring system using Python that can monitor students via their laptop’s webcam and microphone, enabling teachers to observe multiple students simultaneously. The AI monitoring system will incorporate four vision-based capabilities - gaze tracking, mouth open or close detection, person counting, and mobile phone detection that are integrated using multi-threading to operate cohesively.
To run the programs in this repo, do the following:
- create a virtual environment using the command:
python -m venv venv
- activate the virtual environment
cd ./venv/Scripts/activate
(windows users)source ./venv/bin/activate
(mac and linux users)
- install the requirements
pip install --upgrade pip
(to upgrade pip)pip install -r requirements.txt
Once the requirements have been installed, The programs will run successfully.
Except for the person_and_phone.py
script which requires a model to be downloaded.
Tensorflow>2
OpenCV
sklearn=0.19.1 # for face spoofing.
Allen Y | GitHub | |||
Divina Josy | Github | |||
Elsa Maria Joseph | GitHub |