This project is a real-time object detection web application built using YOLOv8 and Streamlit.
Upload images
Upload videos
Use live webcam feed
It performs deep learning-based object detection and displays annotated results directly in the browser.
This demonstrates deployment of a computer vision model into an interactive web application.
Python
YOLOv8 (Ultralytics)
OpenCV
Streamlit
NumPy
- Image Object Detection
- Video Object Detection
- Real-Time Webcam Detection
- Adjustable Confidence Threshold
- Interactive Web UI
- Fast and Lightweight Inference
User Input (Image / Video / Webcam) → Frame Processing (OpenCV) → YOLOv8 Inference → Bounding Box & Label Rendering → Streamlit Web Display
pip install -r requirements.txt
Or manually:
pip install ultralytics opencv-python streamlit numpy
streamlit run app.py
The application will open in your browser at:
Image Upload
Video Upload
Webcam
Adjust confidence threshold from sidebar
View real-time annotated results
The YOLOv8 model is loaded once using Streamlit caching.
Frames are processed individually.
Model predicts:
Bounding boxes
Class labels
Confidence scores
Results are rendered dynamically in the web interface.
Smart surveillance systems
Retail analytics
Traffic monitoring
Industrial automation
AI-powered web applications
This project demonstrates:
✔ End-to-end AI application development ✔ Deep learning model integration ✔ Real-time computer vision pipeline ✔ Web deployment skills ✔ User interface integration with AI