Coffee & Viz - Urban Panorama: t-SNE Street Feature Mapping Tool

The Urban Panorama Project is testing different computer vision and machine learning techniques to assess historical and present-day images of streetscapes to investigate urban change. As part of this project faculty members in the Visual Narrative Initiative Frederico Freitas, assistant professor of History, and Todd Berreth, assistant professor of Art + Design, are developing a t-SNE Street Feature Mapping Tool to help researchers use historical streetscape images to understand the spatial distribution of urban features across cities. This tool allows the visualization of clusters of tens of thousands of building facades, or other desired street-level visual features, scraped autonomously from geolocated street-level images. In this talk, Freitas and Berreth will discuss the underlying computer vision and machine learning systems powering the tool and demonstrate the interactive visualization techniques the tool provides.

*Complimentary coffee, tea and light refreshments from local business Anisette will be available beginning at 9:15 a.m. in an adjacent space. To help increase the sustainability of this program, please bring your own coffee mug.