Visual inference optimisation is a term from the fields of artificial intelligence, automation and Industry 4.0. It describes methods with which systems can make smarter, faster and more accurate decisions based on image and video data.
The aim of visual inference optimisation is to improve artificial intelligence so that it can „understand“ images or videos as efficiently as possible and draw conclusions from them automatically – for example, in the quality control of a production line. Here, a camera is used to photograph each product. The AI uses inference to recognise whether the product is flawless or not. If this process is optimised, defects are found even faster and more reliably, and fewer faulty products reach the customer.
A simple example: In a chocolate factory, an AI system checks every piece of chocolate produced. Through visual inference optimisation, the system learns better and better which chocolates are fine and which have small cracks or discolourations. This saves the company time and money – and customers only receive perfect chocolate.
Visual inference optimisation therefore makes AI applications even more powerful and helps companies to improve their processes in a smart and automated way.
Deutsch 
English (UK)