BrainChip Studio adds auto-rotated models, Linux support and an API for easier systems integration

BrainChip Holdings has upgraded its BrainChip Studio, with version 2018.1.

BrainChip Studio is an AI-powered video analysis software suite, delivering ‘high-speed’ object search and facial classification for law enforcement, counter terrorism and intelligence agencies. New features of the 2018.1 release have been added to make it easier to find objects from a variety of camera views, enable large-scale Linux deployments, and add an API that ‘simplifies’ integration with other applications.

The updated software auto-generates rotated models and its one-shot object training, which BrainChip Holdings describes as a ‘unique characteristic of spiking neural networks’, creates a spiking neural network model of an object in its initial captured orientation.

With the new auto-rotation feature, BrainChip Studio will automatically create multiple rotated models, designed to improve object location in other camera views, where the orientation may vary depending on the installation.

Studio 2018.1 also runs on Linux, which BrainChip Holdings says will offer ‘cost-effective, large-scale’ server installations. “As an added benefit,” Bob Beachler, BrainChip’s Senior Vice President of Marketing and Business Development, says: “the streamlined nature of the Linux OS provides a 10% performance improvement for BrainChip Studio, compared to Windows.”

The 2018.1 release also includes a version of BrainChip Studio that can be controlled with an API-level interface. This enables the functionality of BrainChip Studio to be integrated into third party applications, either off-the-shelf video surveillance software or end-users’ custom software. This version of BrainChip Studio can be re-sold by OEM customers or integrated into cloud-based SaaS frameworks.

Bethan Grylls