https://github.com/cj-mills/unity-deep-learning-image-preprocessor
This Unity package is designed for preparing image input to perform inference with deep learning models. The package includes shaders and compute shaders for various image processing tasks, such as cropping, normalizing, and flipping images.
https://github.com/cj-mills/unity-deep-learning-image-preprocessor
Last synced: 3 months ago
JSON representation
This Unity package is designed for preparing image input to perform inference with deep learning models. The package includes shaders and compute shaders for various image processing tasks, such as cropping, normalizing, and flipping images.
- Host: GitHub
- URL: https://github.com/cj-mills/unity-deep-learning-image-preprocessor
- Owner: cj-mills
- Created: 2023-04-04T21:50:11.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-05-04T22:25:01.000Z (over 2 years ago)
- Last Synced: 2025-01-12T13:28:58.847Z (9 months ago)
- Language: C#
- Homepage: https://christianjmills.com/posts/unity-deep-learning-image-preprocessor-walkthrough/
- Size: 53.7 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Deep Learning Image Preprocessor
Deep Learning Image Preprocessor is a custom Unity package designed for preparing image input to perform inference with deep learning models. The package includes shaders and compute shaders for various image processing tasks, such as cropping, normalizing, and flipping images.
## Features
- Crop image to a specific region
- Normalize image color channels based on mean and standard deviation values
- Flip image along the x-axis
- GPU-accelerated processing
## Demo Projects
| GitHub Repository | Description |
| ------------------------------------------------------------ | ---------------------------------------------------------- |
| [barracuda-image-classification-demo](https://github.com/cj-mills/barracuda-image-classification-demo) | Perform image classification using computer vision models. |
| [barracuda-inference-yolox-demo](https://github.com/cj-mills/barracuda-inference-yolox-demo) | Perform object detection using YOLOX models. |
| [barracuda-inference-posenet-demo](https://github.com/cj-mills/barracuda-inference-posenet-demo) | Perform 2D human pose estimation using PoseNet models. |
## Code Walkthrough
* [Code Walkthrough: Unity Deep Learning Image Preprocessor Package](https://christianjmills.com/posts/unity-deep-learning-image-preprocessor-walkthrough/)
## Getting Started
### Prerequisites
- Unity game engine
### Installation
You can install the Deep Learning Image Preprocessor package using the Unity Package Manager:
1. Open your Unity project.
2. Go to Window > Package Manager.
3. Click the "+" button in the top left corner, and choose "Add package from git URL..."
4. Enter the GitHub repository URL: `https://github.com/cj-mills/unity-deep-learning-image-preprocessor.git`
5. Click "Add". The package will be added to your project.
For Unity versions older than 2021.1, add the Git URL to the `manifest.json` file in your project's `Packages` folder as a dependency:
```json
{
"dependencies": {
"com.cj-mills.deep-learning-image-preprocessor": "https://github.com/cj-mills/unity-deep-learning-image-preprocessor.git",
// other dependencies...
}
}
```
## License
This project is licensed under the MIT License. See the [LICENSE](Documentation~/LICENSE) file for details.