Develop Generative Urban Design Tool with AI-Powered Pattern Analysis (GAN-based)
Budget: $250 – $750 USD
I am developing a Master's thesis project in urban design that leverages artificial intelligence to explore how urban layouts respond to the insertion of new projects.
The goal is to build a generative deep learning model using GANs (Pix2Pix or CycleGAN preferred) to generate a new urban pattern in response to a proposed project, such as a square, park, or building, placed within an existing urban fabric.
But beyond generation, the tool must also analyze and evaluate the spatial impact of the generated pattern — measuring how the project affects the structure, density, connectivity, and overall behavior of the surrounding fabric.
The model must not simply attach the new project into the layout but learn to integrate it in a way that reshapes and reorganizes the surrounding pattern intelligently and contextually.
Deliverables:
Dataset Preparation: Organize and preprocess paired image data (before/after)
GAN Model Training: Train on urban pattern changes using image-to-image translation
Interactive Interface: A user-friendly UI for uploading scenarios and visualizing outputs
Automated Urban Analysis: Extract key spatial metrics:
Built Density
Block Massing
Built-to-Void Ratio
Network Connectivity / Integration Visual Control Metrics Urban Compatibility Index (UCI): A scoring algorithm that summarizes how well the new project harmonizes with the existing urban system, structurally and spatially
Ideal Candidate:
Hands-on experience with GAN models (especially Pix2Pix / CycleGAN) Strong Python skills (image processing, ML models) Familiarity with urban or spatial data (even basic understanding) Ability to develop simple web-based UI (Gradio, Streamlit, or equivalent) Clear documentation and communication.
This project is part of a formal academic thesis and will be used for both demonstration and evaluation. You may be credited in the thesis if you wish.
Please include links to past projects involving generative models or urban/spatial work. Proposals without proven GAN experience will not be considered.
The goal is to build a generative deep learning model using GANs (Pix2Pix or CycleGAN preferred) to generate a new urban pattern in response to a proposed project, such as a square, park, or building, placed within an existing urban fabric.
But beyond generation, the tool must also analyze and evaluate the spatial impact of the generated pattern — measuring how the project affects the structure, density, connectivity, and overall behavior of the surrounding fabric.
The model must not simply attach the new project into the layout but learn to integrate it in a way that reshapes and reorganizes the surrounding pattern intelligently and contextually.
Deliverables:
Dataset Preparation: Organize and preprocess paired image data (before/after)
GAN Model Training: Train on urban pattern changes using image-to-image translation
Interactive Interface: A user-friendly UI for uploading scenarios and visualizing outputs
Automated Urban Analysis: Extract key spatial metrics:
Built Density
Block Massing
Built-to-Void Ratio
Network Connectivity / Integration Visual Control Metrics Urban Compatibility Index (UCI): A scoring algorithm that summarizes how well the new project harmonizes with the existing urban system, structurally and spatially
Ideal Candidate:
Hands-on experience with GAN models (especially Pix2Pix / CycleGAN) Strong Python skills (image processing, ML models) Familiarity with urban or spatial data (even basic understanding) Ability to develop simple web-based UI (Gradio, Streamlit, or equivalent) Clear documentation and communication.
This project is part of a formal academic thesis and will be used for both demonstration and evaluation. You may be credited in the thesis if you wish.
Please include links to past projects involving generative models or urban/spatial work. Proposals without proven GAN experience will not be considered.