Einstein Vision product recognition is based on object detection, which identifies individual objects within an image and determines their position and classification. For shelf-analysis scenarios, the functionality depends on enabling the object-detection capability, providing suitable image data for recognition, and integrating that capability into the Planogram Check workflow used by field representatives.

Einstein Object Detection models rely on representative training images containing the objects that the system must recognize. Salesforce ' s object-detection architecture uses labeled images and object-location information to create a model capable of identifying multiple objects within a photograph. The quality and relevance of the image set are therefore important to reliable product recognition.

Within Consumer Goods Retail Execution, the operational component is the Planogram Check. Planogram assessment tasks provide the visit-level process through which representatives compare observed shelf conditions with the expected planogram and capture the resulting compliance information. Assessment Task Definitions define these repeatable field activities and allow them to be incorporated into Action Plan Templates used during visits.

The combination of object detection, appropriate image information, and a defined Planogram Check provides the technical and operational structure required for automated shelf-product detection in the implementation described.

Study Guide Topics:Platform - > Einstein Vision - > Object Detection - > Planogram Check - > Assessment Task Definition

===============