ONNX_Runtime_Web_Yolov8-seg_Batching

This repository demonstrates how to use ONNX Runtime to run Yolov8-seg models in the browser, including support for batched image processing. The example application displays several images and applies the Yolov8-seg model to detect objects and segment them, with results displayed directly on the webpage.

Features

• Object Detection: Detects various objects in images using YOLOv8.
• Segmentation: Provides segmentation masks for specific objects like persons.
• Real-time Processing: Processes images in real-time directly in the web browser.
• Interactive Gallery: Displays processed images with interactive hover effects.

Technologies Used

• ONNX Runtime: For running the YOLOv8 and segmentation models.
• TensorFlow.js: For image preprocessing and manipulation.
• OpenCV.js: Used for drawing masks.
• HTML/CSS/JavaScript: Frontend development technologies.

Before Segmentation

This image shows the state before segmentation.

After Segmentation

This image shows the segmentation results of the web application.

• Installation

Simply clone the repository and open index.html in live server in a web browser .

git clone https://github.com/shimaamorsy/ONNX_Runtime_Web_Yolov8-seg_Batching.git