Machine vision integration and embedded vision engineering for a next-generation access system, from video quality and matching performance to wired and wireless control-system links.
Anonymised client
The client operates in face-recognition technology for access-oriented environments, including ticketing, credentialing, and frictionless workflows. Estigiti was brought in where concept-level promise had to turn into a more credible edge-device path.
The brief was to narrow the platform decision, adapt the CV/ML pipeline to edge targets, and support integration choices early enough that the programme did not drift into expensive guesswork. The work also needed to support both greenfield and brownfield environments.
Figuring out which edge platforms made sense, how a working CV/ML pipeline should be moved onto them, and how the resulting system would connect into existing or newly built environments without becoming fragile.
Estigiti supported the client through a consulting focused on technical direction. That included recommendations on edge platforms and devices, porting of the CV/ML pipeline, R&D around pipeline architecture and model quantisation, benchmarking work, and embedded systems consulting around integration paths.
That made the engagement particularly useful as a transition layer between an algorithmic capability and a deployable device direction. Instead of treating the work as model tuning in isolation, the scope stayed tied to target platforms, system fit, and the practical realities of integrating face biometrics into access workflows.
If you are still deciding between platforms, integration paths, or rollout options, we can help you narrow the technical direction before delivery risk expands.