Coder Social home page Coder Social logo

creativeikep / selfiesegmentationbarracuda Goto Github PK

View Code? Open in Web Editor NEW
45.0 4.0 4.0 321 KB

SelfieSegmentationBarracuda is a human segmentation neural network that runs the Mediapipe Selfie Segmentation on the Unity Barracuda.

License: Apache License 2.0

C# 82.50% ShaderLab 17.50%
deep-learning neural-network unity human-segmentation segmentation barracuda mediapipe selfie-segmentation

selfiesegmentationbarracuda's Introduction

SelfieSegmentationBarracuda

demo_segmentation demo_virtual_background

SelfieSegmentationBarracuda is a human segmentation neural network that works with a monocular color camera.

SelfieSegmentationBarracuda is a Unity Package that runs the Mediapipe Selfie Segmentation on the Unity Barracuda.

Install

SelfieSegmentationBarracuda can be installed from npm or GitHub URL.

Install from npm (Recommend)

SelfieSegmentationBarracuda can be installed by adding following sections to the manifest file (Packages/manifest.json).

To the scopedRegistries section:

{
  "name": "creativeikep",
  "url": "https://registry.npmjs.com",
  "scopes": [ "jp.ikep" ]
}

To the dependencies section:

"jp.ikep.mediapipe.selfiesegmentation": "1.0.1"

Finally, the manifest file looks like below:

{
    "scopedRegistries": [
        {
            "name": "creativeikep",
            "url": "https://registry.npmjs.com",
            "scopes": [ "jp.ikep" ]
        }
    ],
    "dependencies": {
        "jp.ikep.mediapipe.selfiesegmentation": "1.0.1",
        ...
    }
}

Install from GitHub URL

SelfieSegmentationBarracuda can be installed by adding below URL on the Unity Package Manager's window

https://github.com/creativeIKEP/SelfieSegmentationBarracuda.git?path=Packages/SelfieSegmentationBarracuda#v1.0.1

or, adding below sentence to your manifest file(Packages/manifest.json) dependencies block. Example is below.

{
  "dependencies": {
    "jp.ikep.mediapipe.selfiesegmentation": "https://github.com/creativeIKEP/SelfieSegmentationBarracuda.git?path=Packages/SelfieSegmentationBarracuda#v1.0.1",
    ...
  }
}

Usage Demo

Below code is the demo that segment the prominent humans in the scene. Check "/Assets/Sample/SegmentationSample/Script/SegmentationVisuallizer.cs" and "/Assets/Sample/SegmentationSample/Scenes/SegmentationSample.unity" for SelfieSegmentationBarracuda usage demo details.

using UnityEngine;
// Import SelfieSegmentationBarracuda package
using Mediapipe.SelfieSegmentation;

public class <YourClassName>: MonoBehaviour
{
  // Set "Packages/SelfieSegmentationBarracuda/ResourceSet/SelfieSegmentationResource.asset" on the Unity Editor.
  [SerializeField] SelfieSegmentationResource resource;

  SelfieSegmentation segmentation;

  void Start(){
      segmentation = new SelfieSegmentation(resource);
  }

  void Update(){
      Texture input = ...; // Your input image texture

      // Predict segmentation by neural network model.
      segmentation.ProcessImage(input);
      // Segmentation results can be obtained with `SelfieSegmentation.texture`.
      Texture result = segmentation.texture;
  }

  void OnApplicationQuit(){
      // Must call Dispose method when no longer in use.
      segmentation.Dispose();
  }
}

Demo Image

  • The video for demo scene was downloaded from here.
  • The image for virtual background demo scene was downloaded from here.

ONNX Model

The ONNX model files have been downloaded from PINTO_model_zoo > 109_Selfie_Segmentation.

Author

IKEP

LICENSE

Copyright (c) 2021 IKEP

Apache-2.0

selfiesegmentationbarracuda's People

Contributors

creativeikep avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar

selfiesegmentationbarracuda's Issues

Make it portrait

How to set the camera for a portrait? I tried to change in WebcamInput script's value of width to 1080*1920. But the composite image looks rotated on real android/ios devices. In unity editor, it works well.

  • Segmentation texture size returning processed texture 1080*1920 proper resolution. I have debugged all input texture and composition-processed texture. Returning proper resolution. But somehow composition raw image texture looks not proper, camera view looks rotated.

Any solution for this android/ios portrait selfie segmentation? Really helpful if we can do it for portraite screen
IMG_2663

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.