This device is not compatible.

PROJECT


Build an Object Detection Web App with YOLOv8 and Streamlit

In this project, we’ll design and implement an object detection web application using YOLOv8, OpenCV, and Streamlit. With this web app, we will be able to recognize and identify objects in a video.

Build an Object Detection Web App with YOLOv8 and Streamlit

You will learn to:

Detect objects in videos.

Process videos in Python.

Build a basic web interface using Streamlit.

Load and use a pretrained object detection model.

Skills

Artificial Intelligence

Application Deployment

Machine Learning

Computer Vision

Prerequisites

Intermediate knowledge of Python

Basic understanding of object detection

Basic understanding of image processing

Technologies

YOLO logo

YOLO

OpenCV

Python

Streamlit

Project Description

Object detection models are neural networks capable of locating and classifying objects in an image. In this project, we’ll use YOLOv8 (You Only Look Once version 8)—one of the most popular models for object detection.

We’ll create a simple web application using the web app framework Streamlit and the computer vision library OpenCV. The app enables us to upload a video, identify objects in the frame, and provide information about the detected objects using bounding boxes around the detected objects. We will also incorporate helpful features such as a file uploader, custom class selection, and more.

Project Tasks

1

Introduction

Task 0: Getting Started

Task 1: Load the Pretrained Model

2

Streamlit Interface

Task 2: Set Up the Streamlit App

Task 3: Add a Streamlit Form

Task 4: Add a File Uploader

Task 5: Choose the Objects for Detection

Task 6: Select a Confidence Score

3

Object Detection with YOLOv8 and OpenCV

Task 7: Apply Object Detection on Video Frames

Task 8: Add a Spinner

Task 9: Obtain the Coordinates for Bounding Boxes

Task 10: Add Bounding Box Labels

Task 11: Draw the Bounding Boxes

Task 12: Show Detected Objects

4

Show the Result

Task 13: Save the Processed Video

Task 14: Display the Processed Video

Congratulations!