Ignotum (Latin for “the unknown”) is a poncho that utilises lights and optical effects to stop facial recognition AI’s from reading its wearer’s identity.
The very first pieces of clothing in human history had a mainly protective purpose – be it from harsh climate conditions or injuries. Ignotum has a similar purpose, although from a more modern threat: CCTV cameras with embedded person recognition artificial intelligences.
In close collaboration with an AI expert, we created our own object recognition system, utilising machine learning to teach the algorithm what a human is. With this test setup we started running all kinds of different optical light effects against it, until we found a very simple grid pattern which seemed to work well in breaking it.
As part of the Re-FREAM consortium, Ignotum was realised in close collaboration with the Fraunhofer IZM which was of great help of integrating the electronic components into the poncho type of garment. Utilising their new technology of heat bonding electronics to fabrics and textile cables, we were able to create a fashionable yet practical piece of clothing with embedded light fibres.
In final practical tests we proofed, that the light grid pattern together with the design of the poncho and the moiré effects of the fabrics still worked in confusing the AI from reading its wearer as a human.
Re-FREAM is funded by the European Union’s Horizon 2020 research and innovation program under grant agreement No. 825647