The project aims to design an application for automating background removal image process. Users can simply upload their images and then choose their desired result to download.
U-2-Net is initially trained on the DUST_TR dataset, consisting of images and masks. View examples:
We continue training with personal datasets structured as follows:
The trained model, combined with rembg tool, is used for automatic image background removal.
Remove Background: Automatically removes the background from uploaded images.
For a quick demo, please visit this website: https://bradpt.streamlit.app/. You can upload your images and see the result.
-
Clone the repository: https://github.com/tdp1996/Automatic-Batch-Editing.git
-
Install dependencies:
pip install -r requirements.txt
-
Run the application:
streamlit run main.py -
Open your browser and go to http://localhost:5000
Steps:
- Build Docker image:
docker build -t {image's name} - Build Docker container
docker run -p {port} {image's name}
- U-2-Net: https://github.com/xuebinqin/U-2-Net.git
- rembg: https://github.com/danielgatis/rembg.git
- Train your custom model: danielgatis/rembg#193 (comment)
ABR is still a work in progress and in alpha. While it is slowly getting more to where we want it, it is going to take quite some time to have every possible feature we want to add. If you are knowledgeable in the field of image processing, feel free to contribute to this project and help us get closer to that goal.
- Please fork to your Git account or clone directly the repository Automatic-Background-Removal
- Open a pull request and await feedback from our development team.
If you find a bug or have a feature request, please open an issue and provide detailed information.
- Name: tdp1996
- Email: [email protected]
This project is licensed under the MIT License.