Walking Milano: Unveiling the City’s Character Through 360° Street-Level Panorama Imagery. (sotm2025)

Between September 2024 and August 2025, we conducted a comprehensive street-level survey of Milano, Italy, capturing approximately one million 360° panoramic images using a monopod-mounted camera setup. These images were uploaded to Mapillary, contributing to open-access urban geospatial data. This presentation shares practical insights into continuous data collection methods and analyzes urban characteristics discernible from the imagery, such as graffiti prevalence, urban greenery distribution, and the potential of these visuals as foundational data for 3D digital twin models. I will discuss the current capabilities and limitations of using crowdsourced street-level imagery for urban analysis and planning. Between September 2024 and August 2025, we undertook a comprehensive street-level survey of Milano, Italy, capturing approximately one million 360° panoramic images using a monopod-mounted camera setup. These images were uploaded to Mapillary, contributing to open-access urban geospatial data. This presentation shares practical insights into continuous data collection methods and analyzes urban characteristics discernible from the imagery, such as graffiti prevalence, urban greenery distribution, and the potential of these visuals as foundational data for 3D digital twin models. I will discuss the current capabilities and limitations of using crowdsourced street-level imagery for urban analysis and planning. Advancements in consumer-grade 360° cameras have significantly enhanced image resolution, with modern devices achieving up to 11K. These improvements, coupled with enhanced low-light performance, have expanded the temporal window for effective data collection beyond daylight hours. Mapillary’s object detection capabilities can identify over 150 object classes, including traffic signs, poles, and vegetation. However, challenges remain in detecting certain features like graffiti and nuanced vertical urban greenery, highlighting areas for future development. By analyzing the Milano dataset, we can assess the efficacy of current detection algorithms and identify gaps where manual annotation or algorithmic refinement is necessary. This analysis informs strategies for leveraging 360° imagery in urban planning, such as monitoring infrastructure conditions and informing greening initiatives. The session will conclude with a discussion on the future of crowdsourced 360° street-level imagery, exploring how community-driven data collection can support comprehensive urban analysis and planning efforts. Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ about this event: https://2025.stateofthemap.org/sessions/NRJEVM/