Cogisen

Applied Engineering
for AI

Cogisen is applying a collective knowledge of neuroscience, physics, mathematics and computer science to develop unique algorithms to help AI and machine learning applications extract context and temporal information from content.

Cogisen developed a patented technology to isolate the frequency domain signals containing the relevant information. Our product is the Sencogi® library. The entire toolchain required for this platform is proprietary Cogisen software created in-house.

Applications
of Temporal Models

While current technologies process video as a sequence of individual still images, SENCOGI® uses dynamic motion and sequencing to extract and combine information that is only found only found in temporal effects.

Cogisen’s technology can operate as software-only and plug into current server, mobile or programmable hardware platforms delivering more accurate results faster.

Applications for SENCOGI® include:
• • A motion information model for codecs
• • A software-only gaze tracking solution
• • The detection of violence in IoT applications operating video streams
• • Intelligent AI automating on the monitoring of networks for cybersecurity threats

The
Cognitive Modelling Lab

The focus of Cogisen’s Laboratory in Rome is to deliver commercial AI products that are beyond the current state-of-the-art technology to extract context and temporal information from visual and other inputs.

Cogisen has been able to create cognitive models relying on a probabilistic space, defined by normalized complex vectors. This allows for temporal effects to be efficiently identified as probabilities assigned to histories or a sequence of events.

In the future the Lab will examine different applications of the platform with its technology partners to optimise its adoption, as well as undertaking academic research with leading universities on future innovations in cognitive modelling.