Oss ← Forensic Architecture

Search

FAOpen Source Software

As a research agency, we are committed to creating ways to generalise and democratise the methodologies we use during investigations. As much as possible, we release code, models, and assets so that our techniques can be used beyond FA’s work alone. All of our software and assets are licensed with the permissive Do No Harm license, and many are used and actively co-developed by other research groups in human rights, the art world, and open source journalism.

Different assets have different relationships to internal imperatives, but all are intended as a way to make our work more open, accessible, and collaborative. Some studies here are independent research and development initiatives, whereas others are projects that derive from work done in particular investigations.

All codebases available on Github

study Using Timemap for Cartographic Platforms

Date

Timemap and datasheet-server work together to interactively serve geospatial events from a Google Sheet. The sheet must be formatted appropriately for this to work. This tutorial will walk you through setting up for this example Google Sheet.

study Detecting Tear Gas: Vision and Sound

Date

This study documents several case studies undertaken by Forensic Architecture researchers and OSS contributors. Following on from research we released in early 2019, in the course of the year we have expanded our research in human rights and machine learning, with a particular focus on applications of synthetic data.

study Computer Vision in Triple Chaser

Date

For our contribution to the 2019 Whitney Biennial at New York’s Whitney Museum of American Art, we developed a machine learning and computer vision workflow to identify tear gas grenades in digital images. We focused on a specific brand of tear gas grenade: the Triple-Chaser CS grenade in the catalogue of Defense Technology, which is a leading manufacturer of ‘less-lethal’ munitions.

study Experiments in Synthetic Data

Date

In many investigative contexts, the abundance of potentially relevant data presents a significant challenge to constructing a case. To this end, we use machine learning tools where appropriate to channel investigative resources effectively when processing footage from YouTube, Vimeo, or other media sources.