Alongside the products, we run a strand of academic research applying computer vision to aerial and street-level imagery. The question underneath all of it is the same: how do you turn what a place looks like into something you can measure, compare, and argue with?
Capture imagery systematically, register it to a common frame, then classify it consistently enough that change over time is a measurement rather than an impression.
A per-lot timeline of two New Orleans neighbourhoods after Hurricane Katrina. Historical satellite imagery is warped to a common reference frame so each lot can be classified at every available date, and the recovery trajectories of the two areas compared directly.
Street-level imagery analysed across a small city for quality of life, heat, and equity — turning what a street looks like into measures a planner can compare between blocks.
Structured capture from routine field visits combined with continuous street-level coverage, building the proprietary datasets that computer vision and operational intelligence depend on.
This work is academic and ongoing, and is not distributed as a product. If you are working on something related, or want to talk about the method, get in touch.