Skip to content

A practical approach by implementing the natural image segmentation model for coherent structure and object identification in PIV/PTV fluid experiments

License

Notifications You must be signed in to change notification settings

AliRKhojasteh/Flow_segmentation

Repository files navigation

Flow_segmentation

This is the official repository for "Segment Anything in Flow Experiments".

News

  • 2024.04.15: First release
Open In Colab

Installation

  1. Create a virtual environment: conda create -n flowsam python=3.10 -y and activate it: conda activate flowsam.
  2. Clone the repository: git clone https://github.com/AliRKhojasteh/Flow_segmentation.
  3. Enter the Flow_segmentation folder: cd Notebook.

Get Started

Follow the instructions inside the Notebook folder. Open the Flow_segmentation.ipynb. Within the notebook, you will automatically install and clone required packages, input your image and the text prompt. The model checkpoints are downloaded and available.

  1. Click on "Open in Colab" if you would like to run on your browser without the need for further installations.
  2. Read dependencies and install them
  3. Loading the Input Image "Fingers.png"
  4. text_prompt = 'Fingers and a hand'
  5. Compute masks

*2D PIV experiment of a moving hand, textual input = fingers + hand*

Sample Usage Permissions

All examples available in the 'demo' directory are permitted for demonstration purposes only. For additional usage permissions, please contact the corresponding authors listed in the references.

References

This project makes use of the following repositories:

Citation

@article{khojasteh2024practical,
  title={Practical Object and Flow Structure Segmentation using Artificial Intelligence},
  author={Khojasteh, Ali Rahimi and van de Water, Willem and Westerweel, Jerry},
  note={Submitted to Experiments in Fluids},
  year={2024}
}

About

A practical approach by implementing the natural image segmentation model for coherent structure and object identification in PIV/PTV fluid experiments

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published