PopularMajor Project
Artificial Intelligence
Face Recognition Attendance System
Camera-based attendance with liveness detection and analytics
- Python
- OpenCV
- FaceNet
- Flask
- +2
₹4,900₹5,999
4.8/301 reviews
Artificial Intelligence · Capstone
Abnormal Activity & Loitering Detection in CCTV
Spots fights, falls and loitering in surveillance video
5-6 weeks build
Overview
A surveillance analytics pipeline that tracks people, extracts pose keypoints over short clips and classifies the motion as normal, fight or fall with a temporal model. Loitering is caught separately by dwell time inside operator-drawn zones, and every alert carries a clip so a guard can confirm it before acting.
What makes it stand up
- 01Pose-sequence classifier for fights and falls over sliding windows
- 02Dwell-time loitering rules inside zones drawn on the camera view
- 03Trained and evaluated on public clips such as UCF-Crime
- 04Alert clips saved for guard confirmation, not automatic action
- 05Per-class precision, recall and false-alarm rate in the report
Module breakdown
- 01
Tracking
Person detection with persistent IDs across frames.
- 02
Pose features
MediaPipe keypoints normalised into motion sequences.
- 03
Activity model
Temporal network classifying each track window.
- 04
Zone rules
Polygon zones with dwell thresholds for loitering.
- 05
Alert console
Flask dashboard with clip playback and acknowledgement.
After this, you will be able to
- Explain why pose sequences generalise better than raw pixels
- Discuss false alarms and how thresholds trade them off
- Demo a fall and a loitering alert on recorded footage
Technology stack
- Python
- PyTorch
- OpenCV
- MediaPipe
- Flask
- MongoDB
- Level
- Capstone
- Domain
- Artificial Intelligence
- Build time
- 5-6 weeks
- Difficulty
You receive
- Complete, commented source code
- 60-90 page project report (IEEE format)
- Editable presentation deck
- Architecture, ER and UML diagrams
- Local setup and deployment guide
- Sample dataset or seed data
- Viva question bank with answers
- One 45-minute walkthrough call
- 30 days of doubt-clearing support