Python • OpenCV • MobileNetSSD
AI Vision Detector
A computer-vision project built with Python, Flask, and OpenCV. Upload an image or use your camera to explore object detection in the browser.
Original implementation
Python-powered real-time detection
The original project uses Flask to serve an MJPEG camera stream, OpenCV's DNN module to load a MobileNetSSD Caffe model, and NumPy to process frames. It detects 20 everyday object classes at a confidence threshold above 20%.
Pipeline
- 01 Capture frames with OpenCV
- 02 Resize into a 300 × 300 DNN blob
- 03 Run MobileNetSSD inference
- 04 Draw labels and bounding boxes
Project imagery
Scenes ready for detection
Try one in the browser preview below.
Loading AI model...