Courseiva

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

What is 'image embedding' in computer vision and how is it used in visual search?

⚠ Common exam trap

A common mix-up: candidates confuse 'embedding' as a general computing term (e.g., embedding an object in a document) with the specific machine learning concept of vector embeddings that capture semantic meaning for similarity search.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

Converting images to vectors that capture visual meaning for similarity search and retrieval

Image embedding converts images into dense vector representations (embeddings) that capture semantic visual features such as shapes, colors, and textures. In visual search, these embeddings enable similarity comparisons by calculating distances (e.g., cosine similarity) between query image vectors and a pre-indexed database of image vectors, allowing retrieval of visually similar images even without textual metadata.

Answer analysis

Option-by-option breakdown

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

  • Inserting an image into a Word document or web page as an embedded object

    Why it's wrong here

    Inserting an image into a document or web page creates an OLE object or an <img>/<picture> element that references the binary image data; this everyday use of the word 'embed' means physically placing content inside a file. It performs no mathematical transformation of the image and produces no vector representation. The machine-learning sense of embedding is entirely different: a dense numerical array computed by a model so that similar images have similar vectors, enabling algorithms to search by visual similarity.

  • Converting images to vectors that capture visual meaning for similarity search and retrieval

    Why this is correct

    Image embeddings are produced by passing an image through a trained neural network (such as a convolutional or vision-transformer model) that maps the pixel content into a high-dimensional numeric vector. In that vector space, images that are visually or semantically similar land closer together, which is why distance measures like cosine similarity power reverse image search, product matching, and duplicate detection. This is a learned representation, not a file operation.

  • Compressing images before embedding them in a database to reduce storage costs

    Why it's wrong here

    Image compression (JPEG, PNG, WebP, etc.) is an encoding algorithm that removes redundant or less perceptible data to reduce the number of bytes needed to store the pixel matrix. Even after compression, the database still stores an image file, not a semantic vector; the operation says nothing about visual meaning or relationships between images. An image embedding is a model-generated list of numbers representing learned visual features, which can be indexed with approximate nearest-neighbor search for retrieval, unlike compressed image blobs.

  • Annotating images with GPS coordinates embedded in the file metadata

    Why it's wrong here

    GPS coordinates written into EXIF/IPTC metadata are literal geographic facts (latitude and longitude) about where the photo was taken; they are not generated by a machine-learning model and carry no information about the visual content of the image. Image embeddings, by contrast, are abstract numeric vectors that a model learns to encode the appearance and semantics of objects, scenes, and styles. Although GPS tags enable location-based filtering or map displays, they cannot support content-based similarity matching such as finding 'visually similar' product photos.

About these practice questions

This AI-900 question is part of Courseiva's 985-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

JA

Written by Johnson Ajibi, MSc IT Security

Senior Network & Security Engineer · founder of Courseiva

This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.