Objective:

This assignment aims to familiarize you with object detection techniques using OpenCV.

You will implement a program that can detect and identify objects in videos.

Task:

Write a python code that detect the objects in the $$road$\_$trafifc.mp$4$ video.

$$ Download YoloV$3$ pre$-$trained model from the OpenCV package.

$$ Use OpenCV functions to read and load the pre$-$trained model.

$$ Familiarize yourself with the model's configuration parameters and adjust them

as needed.

$$ Define the class labels associated with the model's predictions.

$$ Convert the captured frames to the format expected by the model.

$$ Perform any necessary pre$-$processing steps $($e$.$g$.,$ resizing, normalization$).$

$$ Pass the pre$-$processed frame to the model for inference.

$$ Parse the model's output, which includes bounding boxes and class predictions.

$$ Draw bounding boxes around detected objects in the frame.

$$ Display the predicted class label and confidence score for each detection.

$$ Display the annotated frame in real$-$timeObjective:

This assignment aims to familiarize you with object detection techniques using OpenCV.

You will implement a program that can detect and identify objects in videos.

Task:

Write a python code that detect the objects in the $$road$\_$trafifc.mp$4$ video.

$$ Download YoloV$3$ pre$-$trained model from the OpenCV package.

$$ Use OpenCV functions to read and load the pre$-$trained model.

$$ Familiarize yourself with the model's configuration parameters and adjust them

as needed.

$$ Define the class labels associated with the model's predictions.

$$ Convert the captured frames to the format expected by the model.

$$ Perform any necessary pre$-$processing steps $($e$.$g$.,$ resizing, normalization$).$

$$ Pass the pre$-$processed frame to the model for inference.

$$ Parse the model's output, which includes bounding boxes and class predictions.

$$ Draw bounding boxes around detected objects in the frame.

$$ Display the predicted class label and confidence score for each detection.

$$ Display the annotated frame in real$-$time