AI-900 Practice Question: Describe features of computer vision workloads on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Scanning product barcodes to look up inventory information

  Why it's wrong here

  Barcode scanning does not leverage computer vision for product identification; it simply decodes a universal product code (UPC) and uses a database lookup to retrieve inventory details. In contrast, product recognition uses machine learning models to visually analyze product appearance, such as color, shape, and label graphics, without relying on any encoded identifier. Since the question specifically asks about identifying products using computer vision, barcode scanning is a traditional, non-vision technique and therefore incorrect.

• ✓

  Identifying retail products and checking shelf placement compliance using computer vision

  Why this is correct

  This is the correct answer because product recognition in Azure AI (e.g., Custom Vision or Azure AI Vision Image Analysis) is designed to detect and label retail products from shelf images, allowing automated assessment of planogram compliance. The model identifies each product by its visual features and can compare detected placement against the expected layout, indicating whether items are out of stock, misaligned, or incorrectly placed. This directly matches the scenario of using computer vision for retail product identification and shelf placement verification.

• ✗

  Generating product descriptions from images for e-commerce listings

  Why it's wrong here

  Generating product descriptions from images is a generative AI task, often accomplished with multimodal large language models that take an image as input and synthesize coherent natural-language text. Product recognition, in contrast, is a discriminative task that outputs product identities and bounding-box locations, not generated descriptions. The underlying models and outputs are fundamentally different: recognition involves classification/detection, whereas description generation involves language generation from visual inputs, making this option incorrect for the stated product-recognition scenario.

• ✗

  Detecting counterfeit or damaged products in a manufacturing quality line

  Why it's wrong here

  Detecting counterfeit or damaged products on a manufacturing line is a quality-control task that typically uses defect detection or anomaly detection algorithms, such as inspecting surface textures, cracks, or irregularities. Product recognition, however, is specifically scoped to identifying and localizing retail products in shelf imagery for merchandising analytics, not for evaluating physical condition or authenticity. Although both use computer vision, the goal, data, and output differ: one classifies defects in manufactured items, while the other identifies SKU presence and placement in a retail environment.

Go deeper

Related to this question