OpenDataCam

Designing an open source tool for offline, transparent and privacy respecting AI-based data collection in urban environments. Visit the project site.

Company: moovel (move lab)
Collaborators: Benedikt Groß, Thibault Durand, Florian Porada, Raphael Reimann
Appearances: Creative Applications, heise online, NVIDIA, balena

Project

2018 - 2020

Testing the object detection with street footage from New Delhi

The discussion around the use of artificial intelligence in public space is ongoing. To contribute to this discussion we developed OpenDataCam, an open source tool that quantifies and tracks moving objects with live video analysis to better understand interactions in urban environments.

The tool never records any photo or video data. It only saves meta-data, in particular the path an object moved or number of counted objects at a certain point. The novelty of the OpenDataCam is, that everything happens locally, while no visual data is saved or sent to online cloud processing.

OpenDataCam runs on Linux and CUDA GPU enabled hardware. It is optimized for the NVIDIA Jetson Board series. The most affordable setup runs on a Jetson Nano combined with other off-the-shelf equipment like a webcam and power supply. The software contains a friendly user interface and is currently optimised for detecting and counting traffic participants, but not limited to that.