B-Automotive
B-Automotive will revolutionise the passenger experience making the car of the future safer, smarter and more responsive to passenger mood and health.
Our technology can help meet current and planned international safety monitoring requirements, allow manufacturers to develop intelligent cockpits and smart interiors personalising passengers’ journey experience.
Ultimately our ability to detect medically relevant expressed behaviour could be used by the car of the future to detect the early signs of neurodegenerative illness, such as Alzheimer’s or Parkinson’s.
Why use B-Automotive?
Using B-Automotive’s interactive AI driven face and voice analysis modules, you will be able to:
Meet current and planned international safety monitoring requirements
Personalise passengers’ journey experience by being able to respond to both their mood and their health
Detect the early signs of degenerative illness
Future ready
Safety - Beyond Safety Monitoring
Using your in-car camera BlueSkeye’s B-Automotive software can track your driver’s gaze and head movement whilst our computer vision capability combined with artificial intelligence monitors and interprets changes to the facial behaviour assessing their levels of concentration and tiredness.
This information is key to preventing accidents and meeting the developing requirements of international regulators for example Euro NCAP
You can also use the microphones in your vehicle to identify speech and analyse pitch and tone, adding further levels of data and increased precision to that assessment.
Our B-Automotive SDK is hardware-agnostic making it easy to integrate into any model of vehicle and lightweight in terms of both computing and electrical power.
Personalisation
Using your car’s embedded camera(s) and microphones BlueSkeye’s B-Automotive software can help you create a truly “intelligent cockpit” allowing your vehicle to measure and respond to your occupants’, moods, emotions and even the state of their health.
By combining data from your car’s cameras and microphones, BlueSkeye’s B-Automotive software continuously analyses and measures changes to occupants’ facial behaviour, as well as the tone and pitch of the voice to reliably and accurately draw conclusions about their current mood.
You can incorporate this information into your vehicle’s on board computing systems (CANBUS) and fine tune the car’s environment; adjusting the temperature and air conditioning, seat position, in-car entertainment, lighting and ride based on real time passenger feedback allowing each journey to be customised to your passengers’ preferences.
Expressed Emotion with AI
BlueSkeye B-Automotive uses a continuous approach to measure expressed emotion. This better fits the real human experience of emotional states. This approach allows emotion regions to be defined and to measure the transitions away from and towards these regions.
This continuous approach, where appropriate, can be mapped back to a much less exact categorical representation. For example, excited, calm, or angry.
We use in-car cameras to monitor facially expressed emotion using facial muscle movements, Facial Action Units (AUs) gaze behaviour, and the pose of the head. Audio information can also be used but this does not generalise well to noisy multi-occupant environments like automotive cabins. This is an area we are continuing to actively research.
B-Automotive uses a continuous approach to the categorisation of emotion (Apparent VA). This better fits the real human experience of emotional states and avoids having to quantify the fuzzy boundaries and subjective definitions of discrete emotional states between different individuals.
What is B-Automotive compatible with?
B-Automotive is designed to be platform agnostic
The development platform currently supports the following CPU architectures:
● X86
● ARM
on Linux platforms
Acceleration such as GPU is not required but can be used if available.
Automotive SDK requires 250 MB of runtime RAM usage assuming 720p input and 500 MB of disk space usage to store B-Automotive itself, as well as the included machine learning models.
The deployed system is highly configurable and can be optimised for best fit with the available compute.
The system requires approximately 10 seconds to begin generating robust and reliable data.