This project is a collaboration between Segment Anything and YOLOv8 algorithms, focusing on object segmentation. The goal of the project is to automatically identify and segment objects in images, providing region-specific highlights.
Note: Also, if you want to segment low light images by illuminating them, you can refer to my Low-light-segmentation work.
To run this project locally, please follow the steps below:
- Clone this repository:
!git clone https://github.com/SelimSavas/segment-anything-and-yolov8.git
- Install the required dependencies (Do this if you are using colab. If you are working in your external environment, pay special attention to the installations required):
!pip install -r requirements.txt
- Run the project:
To perform image segmentation, you can use the provided Colab notebook: SAMandYOLOv8.ipynb.
For video segmentation, you can utilize the Colab notebook: SAMandYOLOv8video.ipynb.
If you prefer to run the video segmentation on your own local environment, execute the following command:
!python SAMandYOLOv8video.py
Here are some example results from the project:
Image Segmentation
Video Segmentation
result2.online-video-cutter.com.mp4
Contributions to this project are always welcome. If you find any bugs or have suggestions for improvement, please open an issue or submit a pull request.
This project is licensed under the Apache Licene. See the LICENSE file for more information.
We would like to express our gratitude to the creators of Segment Anything and YOLOV8 for their invaluable contributions to the field of computer vision and object segmentation.
For any questions or inquiries, please contact 16savasselim@gmail.com
Feel free to explore and use this project for your own applications. Happy segmenting!
You can throw a star to support.