InvigilAI is an AI-powered online examination monitoring system designed to ensure fair and secure remote assessments. It focuses on detecting suspicious activities during exams and maintaining exam integrity through smart monitoring.
The detections include:
- Multi-Person Detection
- Real-time Identity monitoring
- Talking/Whispering Detection
- Gaze & Iris tracking
- AI-based exam monitoring
- Real-time activity tracking
- Secure authentication (Auth0)
- Clean and responsive UI
- Scalable frontend architecture
This image shows the process of registering an identity for secure access during exams.
Real-time detection of suspicious activity during the exam. Our models constantly monitor user behavior to ensure fair assessments.
System Architecture
- React
- Tailwind CSS
- Auth0
- Python
- YOLO
- PANNs
- MediaPipe
- SK-learn
- VGG2
- Docker
- FaceNet
npm install # install frontend dependencies
npm run dev # start frontend dev servercd InvigilAI/Backend # move into backend directory
docker-compose up --build # build images and start backend servicesBackend/
├── docker-compose.yml
├── .gitignore
│
├── vision_service/
│ ├── main.py
│ ├── vision_module.py
│ ├── gaze_module.py
│ ├── weights/
│ │ └── best.pt
│ ├── requirements_vision.txt
│ └── Dockerfile.vision
│
├── identity_service/
│ ├── main.py
│ ├── identity_module.py
│ ├── enrollment_utils.py
│ ├── data/
│ │ ├── svm_model_facenet.pkl
│ │ └── label_encoder.pkl
│ ├── requirements_id.txt
│ └── Dockerfile.id
│
├── audio_service/
│ ├── main.py
│ ├── audio_module.py
│ ├── panns_data/
│ │ └── Cnn14_mAP=0.431.pth
│ ├── requirements_audio.txt
│ └── Dockerfile.audio
│
└── gateway_fusion/
├── main.py
├── fusion_module.py
├── requirements_gateway.txt
└── Dockerfile.gateway

