Measure the Future

Measure the Future is an open source project that is building a computer-vision based tool to allow libraries, museums, and other civic locations to better understand the use of their physical spaces and buildings. We are building a tool based on Raspberry Pi hardware and an open software stack (Raspian Linux, OpenCV, Go, SqlLite, ReactJS) that provides insights into space usage, but does so with a privacy-first architecture that attempts to prevent any identification of individuals. We have alpha tested our solution and are confident that we are on the right path, and are looking to improve usability, flexibility, and visualizations of the data we collect.

Ideal Candidate

Measure the Future has two pieces that, while separate, have to be kept in connection with each other. The first is the Scout, our sensor package that relies on the hardware component and is built on top of OpenCV using Go. The second is the Mothership, our web-based visualization engine that takes the data provided by the Scout and makes it usable and understandable for the end-user. We are open to developers with either skill set, either OpenCV backend development in Go, or front-end visualization using ReactJS (or other web-standard library). Priority will be given slightly to front-end data analysis and visualization, but we are open to inquiries from anyone interested in the goals of the project.

Project Ideas

UI/UX Enhancements and Visualizations

Our Scout provides a very large amount of data in the form of locational data and time stamps, and there are many ways to make that data interesting and useful. Your job would be to take data from a Scout, and find ways to make it amazing. Build a visualization, an amazing user interface to be able to query the data in a variety of ways, a new way of interacting with our data that amazes and fascinates our users.

OpenCV Enhancements

Computer vision is a complicated and touchy thing, and our system always needs to be better at the basics (identification of people, blob tracking) as well as the much more complicated (identification of adults vs children by size, temporary identification of individuals for deduplication purposes, etc). You'll take a look at our code and make it awesomely better at grabbing the data that we need to answer questions about space for our users.