According to a July 2019 article in The New York Times, dozens of databases of people’s faces are being complied without their knowledge by companies, universities, government entities and others. These databases – some with millions of faces – are sometimes being shared worldwide in a vast and growing facial recognition technology ecosystem.
Rapid expansion of personal recognition databases – and the spread of facial recognition technology itself – has escalated pushback against what is increasingly perceived as an intrusive, privacy-violating technology. This, in turn, has sparked interest in applying new ways of using video camera data effectively, but without facial recognition.
Sense Vision features the breakthrough Sense Vision Positioning System (VPS) for object location and movement tracking without the need for mapping data or instrumenting objects or people being tracked. VPS can determine locations in 3D space using a monocular camera.
As people and objects within camera imagery are annotated for machine learning (ML) model training data sets, VPS automatically calculates each labeled object’s location, and with Vision Depth Perception (VDP), measures the proximity distance of one object in relation to another, as well as the distance the object moves within the camera view over time. No facial recognition is needed nor used, to deliver these capabilities.
Sense Vision is able to preserve and improve the value of cameras in the enterprise without Orwellian facial recognition. The result is always-on, unblinking video camera and visual intelligence that enables live, remote monitoring for use cases such as workforce management, safety, security, inventory management, regulatory compliance and other needs for monitoring and engaging people, spaces and things – without identifying individuals.
And Sixgill provides other options for greater privacy. For example, camera data does not have to be stored on servers or in the cloud. Sense Vision Edge capabilities enable deployment of the camera data artificial intelligence (AI) on or near the camera itself so cameras can respond to what they see in the moment without the forcing data retention. Thus, Sense Vision imagery can be acted upon with the option to not retain it, when privacy safeguards are paramount.
Enabling More Value From Smart Cameras
Video cameras are commonplace because of their ability to perform continuous real-world information gathering. With advancements in connectivity, Edge Computing and Machine Learning, smart cameras can deliver more value to real-time operations. At Sixgill, we encounter many customers for whom the camera is the sensor of first resort. That’s largely because cameras are understood and easy to implement, posing less mystery to buyers than mixed-sensors networks.
According to Gartner, by 2020, 65% of enterprise-captured video/image content will be analyzed at least partially by machines.
Increasingly, organizations are facing difficult challenges to accurately and comprehensively capture this high-intensity data and use it effectively without relying on human monitoring.
Sense Vision with Machine Learning was created to maximize the value of camera data in our customers’ real-time operations. It’s a natural extension to Sense, as video is just another form of time-series data, for which Sense is comprehensively optimized, at whatever frames-per-second rate video is captured.
With Sense Vision, camera data is ingested via the Sense Vision Data Pipeline and prepared for deep learning and analysis. Sense Vision processes and enriches live video data with existing or custom ML classifications, object recognition or entity extraction models.
Fully Automated Camera Data Automation At Scale
With Sense Vision, CIOs, CTOs and other systems builders can capture the full value of their enterprise-wide camera data collection with fully automated, data-driven actions at scale. Custom dashboards make continuous metrics monitoring and system administration easy. Other capabilities include:
- Monitoring and tracking location, volume, movement, and behavior of people and things
- Understanding building, room, and common space utilization
- Detecting objects, identifying activities and behaviors, and triggering real-time actions
- Enforcing location- and situation-based safety, security, and compliance
- Greater privacy safeguards in networked camera deployments
Sense Vision can deliver insights and programmatic responses from video data streams in the Cloud, at the Edge, or both. Edge deployment on the cameras themselves provides reduced processing costs, low-latency immediacy, and greater personal privacy.
Sixgill’s Sense Vision is perfect for enterprises that demand complete ownership of data, freedom from data balkanization, and autonomy from lock-in proprietary contracts. Contact Sixgill to learn more and schedule a demo.