Project Overview This project implements an Object Detection system using Computer Vision and Deep Learning to identify and localize objects in images and live camera feeds. The model detects multiple objects simultaneously by drawing bounding boxes and assigning class labels with confidence scores.

Features Detects multiple objects in a single image Draws bounding boxes with labels and confidence Supports image-based inference Supports live camera (webcam) inference Trained on a custom / standard dataset Uses transfer learning for better accuracy

Live Inference (Webcam Detection) The project supports real-time object detection using a webcam. Live Detection Features: Captures frames from camera Runs object detection on each frame Displays bounding boxes in real-time

Results Successfully detects multiple objects High accuracy on validation data Works in real-time with webcam input

Future Enhancements Improve accuracy with larger datasets Deploy as a web application Add object tracking Optimize for mobile devices

SCREENSHOTS