Location>code7788 >text

YoloDotNet v2.1: a tool for real-time object detection

Popularity：102 ℃/2024-10-10 21:48:22

Projects

YoloDotNet v2.1 is a real-time object detection framework based on C# and .NET 8, designed for object detection in images and videos. It integrates Yolov8 ~ Yolov11 models, enables efficient object detection through and ONNX runtime, and supports GPU acceleration (using CUDA.) YoloDotNet not only supports traditional object detection, but also covers classification, OBB detection, segmentation, and pose estimation, which makes it suitable for a wide range of complex vision tasks.

Technical analysis of projects

YoloDotNet 2.1 is now available and is more powerful than ever! This version builds on the previous "Speed Demon" v2.0 update and adds some exciting new features while keeping everything running smoothly. Compatibility with older versions has been ensured and some tweaks have been made for better object detection performance. Check out the new features:

  • Yolov11 Support:The support of the latest and greatest object detection models provides users with more advanced object detection capabilities.
  • Backward compatibility of Yolov9:You can now switch between Yolov8-v11 versions.
  • Minor optimization:There are a few tweaks here and there for faster object detection, the faster the better!
  • OnnxRuntime update:CUDA and cuDNN are now supported. GPUs will definitely be happy with this!

YoloDotNet v2.1 - faster, smarter and includes more Yolo benefits.


Project and technology application scenarios

YoloDotNet v2.1 has a wide range of application scenarios, including but not limited to:

  • intelligent monitoring: Detect abnormal behavior or objects in surveillance video in real time.
  • automatic driving: Real-time recognition of pedestrians, vehicles and other obstacles in the road.
  • Industrial Inspection: Automated detection of product defects or abnormalities on production lines.
  • Medical Image Analysis: To assist doctors in quickly recognizing lesion areas in medical images.
  • Sports Analysis: Real-time analysis of athletes' movements and postures for training and game analysis.

Project Characteristics

YoloDotNet v2.1 has the following notable features:

  • high performance: With a number of optimizations, YoloDotNet v2.1 reaches new heights of speed and efficiency, especially under GPU acceleration.
  • multifunctionalIt supports a wide range of vision tasks such as classification, object detection, OBB detection, segmentation and pose estimation to meet different application requirements.
  • usability: Provides a concise API and rich sample code , easy for developers to quickly get started .
  • cross-platformNET 8-based and supports multiple operating systems including Windows, Linux, and macOS.
  • Open source and free: Fully open source, users are free to use, modify and distribute.

concluding remarks

YoloDotNet v2.1 not only achieves a major breakthrough in technology, but also provides users with powerful tools to tackle a wide range of complex vision tasks. Whether you are a developer, researcher, or enterprise user, YoloDotNet v2.1 provides you with efficient and reliable solutions. Experience YoloDotNet v2.1 today and start your intelligent vision journey!

Project Address：YoloDotNet GitHub

Installation Guide：

dotnet add package YoloDotNet

take note of: Using GPU acceleration requires CUDA and cuDNN to be installed, ensure compatibility of the ONNX runtime with these components.

The project contains a sample project, the startup file is located in theConsoleDemo/. This file contains the project's entry point for starting and running YoloDotNet's console application.

Overview of the content of the document
using System;
using YoloDotNet;

namespace ConsoleDemo
{
    class Program
    {
        static void Main(string[] args)
        {
            // initialization Yolo boyfriend
            var yolo = new Yolo(@"path\to\");

            // Load Image
            var image = <Rgba32>(@"path\to\");

            // 运行boyfriend检测
            var results = (image, confidence: 0.25, iou: 0.7);

            // Disposal results
            (results);
            (@"path\to\save\");
        }
    }
}
Launch File Function
  • Initializing Yolo Objects: Load the ONNX model.
  • Load Image: Use to load an image.
  • Running Object Detection: Call the Yolo object'sRunObjectDetection method for object detection.
  • Disposal results: Plot the detection result on the image and save it.

3. Introduction to the project configuration file

The YoloDotNet project does not have traditional configuration files (such as the.config maybe.yaml file), but the behavior of the project can be adjusted through configuration options in the code.

Examples of Configuration Options
var yolo = new Yolo(new YoloOptions
{
    OnnxModel = @"path\to\",
    ModelType = ,
    Cuda = true,
    GpuId = 0,
    PrimeGpu = false
});
Description of configuration options
  • OnnxModel: Specifies the path to the ONNX model.
  • ModelType: Specify the model type, e.g.ObjectDetection。
  • Cuda: Whether CUDA acceleration is enabled.
  • GpuId: Specifies the GPU ID to be used.
  • PrimeGpu: Whether to pre-allocate GPU memory.

With these configuration options, the behavior of YoloDotNet can be flexibly adapted in code to different application scenarios.


<Previous Next>