Urban Studio is a powerful and user-friendly toolkit for urban design and analysis. It provides a suite of tools and functionalities to aid in urban environment analysis, from image segmentation for building detection to planform generation and 3D reconstruction. Additionally, the project includes a graphical user interface (GUI) for easy and convenient operation.
-
Image Segmentation: Quickly and accurately segment urban environments to analyze the percentage of buildings in the area.
-
Graphical User Interface (GUI): Urban Studio features an intuitive GUI for easy and efficient interaction with the toolkit.
To use Urban Studio, follow these steps:
-
Clone this repository to your local machine:
git clone https://github.com/H-tr/urban-studio.git
-
Create virtual environment (conda is recommanded)
conda create -n urban-studio -y python=3.10 conda activate urban-studio
-
Install the required dependencies and build the project by running:
pip install -e .
We are using mmcv as dependency, so we need to install the pretrained model and config files
mim download mmsegmentation --config segformer_mit-b4_8xb2-160k_ade20k-512x512 --dest src/models
By default, we are using the segformer trained on ade20k dataset. Feel free to try other models!
-
You should specific your image path and output path in
configs
-
Launch the application by running:
python src/main.py --config configs/segmentation.yaml
For example
python src/main.py --config configs/segmentation.yaml
- Image Segmentation: Open an urban image in the Urban Studio GUI. Click the "Segment" button to perform image segmentation and analyze the building percentage.
For detailed instructions and tutorials on how to use specific features, please refer to the documentation.
For comprehensive documentation and detailed guides on using Urban Studio, please visit our Documentation folder. You can find in-depth information on each feature, as well as usage examples.
We welcome contributions from the community. If you'd like to contribute to Urban Studio, please follow our Contributing Guidelines. Your input and enhancements are greatly appreciated!
Here's a list of planned features and improvements for Urban Studio:
- Implement a feature for urban green space analysis.
- Implement a GUI
- Create tutorials for advanced use cases.
Feel free to contribute to any of these tasks or propose your own!
This project is licensed under the MIT License - see the LICENSE file for details.