From Image Generation to Infrastructure Design: a Multi-agent Pipeline for Street Design Generation

2025-09-05

AI Summary

Low visual rendering efficiency in traditional street design impedes public participation and collaborative decision-making, while existing AI approaches rely heavily on large-scale annotated datasets and struggle with precise spatial editing in complex urban street scenes. This paper introduces the first multi-agent generative framework specifically designed for street infrastructure planning—requiring no domain-specific pretraining and enabling direct, pixel-accurate editing and redesign of bicycle facilities on real-world street imagery. The framework integrates four core modules: lane localization, prompt optimization, generative modeling, and automated evaluation, orchestrating multi-agent collaboration to achieve end-to-end design generation with real-time quality feedback. Experiments demonstrate robust geometric plausibility, visual coherence, and instruction fidelity across diverse urban road scenarios. Our approach significantly accelerates design iteration cycles and enhances realism, offering a scalable technical foundation for participatory urban renewal.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Multiagent Systems: Multiagent Planning

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Realistic visual renderings of street-design scenarios are essential for public engagement in active transportation planning. Traditional approaches are labor-intensive, hindering collective deliberation and collaborative decision-making. While AI-assisted generative design shows transformative potential by enabling rapid creation of design scenarios, existing generative approaches typically require large amounts of domain-specific training data and struggle to enable precise spatial variations of design/configuration in complex street-view scenes. We introduce a multi-agent system that edits and redesigns bicycle facilities directly on real-world street-view imagery. The framework integrates lane localization, prompt optimization, design generation, and automated evaluation to synthesize realistic, contextually appropriate designs. Experiments across diverse urban scenarios demonstrate that the system can adapt to varying road geometries and environmental conditions, consistently yielding visually coherent and instruction-compliant results. This work establishes a foundation for applying multi-agent pipelines to transportation infrastructure planning and facility design.

Problem

• Generating realistic street-design scenarios for public engagement
• Overcoming labor-intensive traditional street design approaches
• Enabling precise spatial variations in complex street scenes

Innovation

• Multi-agent system edits bicycle facilities on street imagery
• Integrates lane localization, prompt optimization, and automated evaluation
• Adapts to varying road geometries and environmental conditions