The widespread availability of mobile phones with high quality cameras means that dramatic events around the world can be captured on video and rapidly shared via social media. The growth of live streaming over the next few years means that such video will soon be available immediately in high volume. Because this video is multi-perspectival, it will tell the story of an event from many different angles rather than a single authoritative view.

While this video has the potential to provide valuable information, its complexity and lack of structure makes it very difficult to analyze.

SITU Research and Carnegie Mellon University have partnered to create a technology platform for organizing and analyzing large volumes of event-based video. This platform contains a variety of tools in various stages of development—including a video archiving system, video synchronization algorithms, geolocation capacity, as well as visual interfaces for exploring video data. This system is media format agnostic and allows synchronization of video even in the absence of metadata in many instances.

This system will be useful for human rights investigators and advocates, fact-finders, journalists, scholars, and ordinary citizens as they seek to better understand and analyze events such as protests and demonstrations, mass government repression, episodes of police brutality, conflict events and disasters.

Global Alignment. Robotics Institute, CMU.

In order to demonstrate the capabilities and potential of this system, we are using the Ukrainian Euromaidan protests as an exemplar in this report. However, the infrastructure we have created would work well for any event in which large volumes of video are available, such as Ferguson, Gezi Park in Turkey, the Arab Spring, the November 2015 Paris terrorist attacks.

It is important to note that video—even in high volume—cannot tell the entire story of an event. As with all other forms of evidence, it must be combined with other available data in order to provide a nuanced and accurate understanding of what has taken place. The goal of this system is to make video evidence from an event rapidly accessible and analyzable to those who need it.