Accurate and timely geospatial travel data is an essential tool in designing effective transport systems that support active, sustainable travel. Councils need to be “guided by data”, both in designing long-term strategic travel initiatives and when directing tactical, in the moment interventions. Our proposal is to deploy cutting-edge deep learning techniques to the existing road traffic CCTV cameras in use by local authorities. Our aim is to convert these existing CCTV installations into smart sensors, capable of capturing and recording rich streams of geospatial data covering every aspect of the operation of the road network. Through using CCTV cameras, we will be able to achieve a previously unprecedented granularity and separation between modes of transport, allowing us to specifically aid councils’ in their Active Travel efforts. This will be achieved without the need to install additional cameras or any specialist hardware.