Camera Vision

Post date: Sep 27, 2015 4:59:8 PM

This week I explored the existing research on camera vision, specifically focusing on programs to count pedestrians. There are a wide variety of solutions to this problem coming out of both the academic and business world. These solutions range from focusing on counting thousands of people in large crowds to counting low volumes of people who cross a specific street or doorway. Both of these kinds of solutions would be useful to determine pedestrian flow and density throughout Venice. Indeed automating our data collection process is critical to the success of this project in future years.

If our GIS layered maps could be linked to real time traffic data (which is clearly possible to do) then our navigational app could be made much more robust. Moving to perpetual automated counting would provide year round accuracy as well as a higher data reliability.

I have established a links document which contains the most useful sources which I found. I will begin to track these in EndNote as soon as possible.

-DS