AIVI

We are supporting features for various use-cases and in different environments. From customers shopping in retail stores to large crowds at events, or other objects and areas of interest in the security and surveillance business. Let’s have look at the image below to find out the different components that this end-to-end solution consists of.

On the left side, you have edge devices with video cameras capturing images or video streams – which are processed in real-time, enabling people or objects detection, and various characteristics such as age and gender. A controlling module can send instructions to devices such as display screens, radio players, or smart lights – which are tailored, based on the inputs it received.

From the back-office, a configuration module gives business users the opportunity to tailor the influencing steering parameters based on the targeted customers. The impact and performance can be measured and reported, allowing us to tweak the desired outcomes to even better future results.

This results in a continuous feedback loop. From observing to influencing behaviors, and performance measurements to allow adjustments in the steering.

The ultimate end-goal would be to even automate this tweaking and steering, and support business-decision makers in optimizing the full end-to-end process.

This is our AI Vision Platform, also called AIVI (pronounced as “Ayvie”). A full overview is shown in the picture below.

Any type of (visual) inputs can be processed, regardless of the hardware or operating systems used. Camera streams, videos, or images can be processed in different formats including night vision – infrared images and high- or low resolutions. Even additional structured and unstructured meta-data can be processed by our platform. A centralized platform like AIVI enables capabilities such as high volume storage of detections, scaling up or out based on the load, service monitoring, and even tailored customizations.

AIVI has a lot of intelligent capabilities that will keep growing as new use cases are added to the library marketplace. From detecting and tracking large amounts of objects – even in dense crowds – to recognizing patterns like group forming or emerging queues at a checkout in a store. And, ultimately, AIVI can initiate triggers that can be configured in a very flexible way.

Processing can be done on the edge devices or in the cloud, depending on the performance and security needs.

Any output is supported, mainly in 3 basic categories:

All these outputs can be delivered in real-time, periodically, or based on configurable events or triggers. Outside the AIVI platform boundaries is your organization with your day-to-day business processes, marketing & sales-, or logistics & planning stakeholders, or business-critical applications such as business intelligence and CRM tools. Computer Vision can be applied in a lot of domains. We have reference cases including Retail, Farming, Industry, Security, and more. Connecting to the AIVI platform is simple, and we can offer support in developing tailored connectors where necessary.

This is our full feature catalog of the AIVI platform roadmap, which will be continuously enhanced and improved over time.