Real-time detection system for:
- Face (Age, Gender, Expression)
- Hand Gestures (Finger Counting)
- Object Recognition.
Powered by: Python • DeepFace • MediaPipe • YOLOv8
-
Face Detection:
- Age (with range adjustment)
- Gender (smart override for accuracy)
- Emotion (Happy, Sad, Angry, etc.)
-
Hand Gesture Detection:
- Finger counting in real-time (1–5)
-
Object Detection:
- Recognizes common real-world items
- YOLOv8 pretrained model for high speed & accuracy
-
Install Python 3.7 – 3.10 Download: https://www.python.org/downloads/release/python-3100/
-
Install Required Libraries:
Run these in CMD:
py -3.10 -m pip install --upgrade pip py -3.10 -m pip install opencv-python py -3.10 -m pip install deepface py -3.10 -m pip install mediapipe py -3.10 -m pip install ultralytics py -3.10 -m pip install numpy
If issues occur:
- Try: pip install opencv-contrib-python
- Or create a venv: python -m venv venv
• Combined Detection (Face + Emotion + Hand + Object): cd "C:\Users\Admin\Desktop\Python AI DeepFace\Object Detector-Age-Gender" py -3.10 age_gender_emotion.py
• Object Detection Only: cd "C:\Users\Admin\Desktop\Python AI DeepFace\Object Detector-Age-Gender" py -3.10 object_detection.py
-
Ensure webcam is connected (default index: 0)
-
For fallback detection: from ultralytics import YOLO model = YOLO("yolov8n.pt")
-
Colors: • Object labels = Yellow • Face boxes = Green • Finger count = Above wrist
- DeepFace: https://github.com/serengil/deepface
- MediaPipe: https://google.github.io/mediapipe/solutions/hands.html
- YOLOv8: https://docs.ultralytics.com
- OpenCV
- NumPy
Made with 💡 by Engr Saad — integrating multiple AI capabilities
into a single, real-time vision solution.
Email: [email protected]
Website: https://engr-saad.com
Facebook: https://www.facebook.com/Engr.Saad.2/
Company WEB: https://triangletech.com.bd/
WhatsApp: +8801904654712
- Voice alerts (based on emotion)
- Data logging (CSV)
- Web interface
- Custom model retraining