metadata
title: Cut Out
emoji: ✂️
colorFrom: gray
colorTo: pink
sdk: docker
pinned: false
license: apache-2.0
short_description: An AI image segmentation demo made with Flask and Mobile SAM
Cut Out
Cut out is a image segmentation implementation project that allows users to upload an image, select a region of interest and generate a cutout image with the selected region. In this project, I used Meta's Segment Anything Model which is state of the art deep learning model in the image segmentation task. To try the demo visit here. To try the Meta's demo try here.
Demo
https://github.com/user-attachments/assets/98ee0372-a640-419a-8580-1ebf100566ad
Tech Stack
- Python
- Flask
- Numpy
- Pillow
- Ultralytics
- Torch
- Torchvision
- Werkzeug
Installation
Github
You can clone this repository by
$ git clone https://github.com/jaisuryaprabu/cut-out.git
$ cd cut-out
Create the virtual environment
$ python -m venv <your_venv_name>
$ .venv/Scripts/activate.ps1 <or> source .venv/Scripts/activate
Export the FLASK_APP
Enter this command
$ export FLASK_APP=main.py
Run the app
$ python -m gevent.pywsgi -b 0.0.0.0:7860 main:app
Contribution
Contributions are warmly welcomed 🤗 To contribute :
- Fork this repository
- Make your changes and commit them
- Explain about the new feature or improvements
- Open a pull request with a description of your changes
Future Works
- Video segmentation