Question

You are building a lightweight vision app in Azure AI Foundry for a retail inventory team. The app must analyze shelf photos and return not only the names of the products it recognizes but also the exact rectangular pixel locations where each product appears, so the team can overlay bounding boxes on the images. Which computer vision capability should your app request?

Accepted Answer

Object detection, which returns labels along with bounding-box coordinates for each detected item

Answer

Image classification, which assigns one or more category labels to the whole image

Answer

Optical character recognition (OCR), which reads printed and handwritten text from images

Answer

Image captioning, which generates a single descriptive sentence for the image