AI-900 Practice Question: Describe features of computer vision workloads on Azure
What is 'dense captioning' in Azure AI Vision v4.0?
⚠ Common exam trap
It's easy for candidates to confuse dense captioning with standard image captioning (Option A), assuming 'dense' simply means a longer or more detailed single caption, rather than recognizing it as a region-specific, multi-caption feature with bounding boxes.
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
✓
Generating multiple region-specific captions each with a bounding box for different image areas
Dense captioning in Azure AI Vision v4.0 goes beyond describing the entire image; it identifies multiple distinct regions within the image and generates a separate caption for each region, along with a bounding box that pinpoints its location. This allows for granular understanding of complex scenes, such as recognizing 'a dog on a couch' and 'a lamp on a table' as separate, localized descriptions.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Generating a very long and detailed caption for the entire image
Why it's wrong here
A single very long caption for the entire image, regardless of its verbosity, still provides only one image-level description with no spatial granularity. Dense captioning specifically detects multiple distinct regions within the image and generates a separate caption for each region, with each caption tied to its corresponding bounding box. The key is the multiplicity and localization of captions, not the length of a single caption. Therefore, this option misses the core concept of region-specific, spatially grounded descriptions.
- ✓
Generating multiple region-specific captions each with a bounding box for different image areas
Why this is correct
Dense captioning in Azure AI Vision detects objects or regions of interest within an image and generates a natural-language caption for each region, returning both the caption text and a bounding box indicating where that region is located. For example, an image with a dog and a bicycle would yield separate captions for the dog and the bicycle, each with its own bounding box coordinates. This region-level granularity provides richer context than a single whole-image caption and enables applications like accessibility, image search, and content analysis to reference specific parts of an image.
- ✗
Adding caption text overlaid on top of the image like movie subtitles
Why it's wrong here
Overlaying caption text on top of an image, like movie subtitles, is an image rendering or video-editing operation that modifies the visual output rather than producing structured analysis data. Dense captioning is an API feature that returns structured JSON containing the caption strings and their associated bounding boxes; it does not generate a modified image with text burned in. Movie subtitles are also time-synchronized spoken-dialogue transcriptions for videos, whereas dense captions describe static visual content across multiple regions simultaneously, so this option conflates a display technique with an AI analysis output.
- ✗
Captions that include technical details like camera settings and lighting conditions
Why it's wrong here
Camera settings and lighting conditions are EXIF metadata recorded by the camera hardware at capture time, not semantic descriptions generated by an AI vision model. Dense captioning analyzes pixel content to produce natural-language descriptions of what is actually visible in different image regions, such as 'a red car parked in front of a store.' It does not read or report technical metadata, so this option confuses embedded image metadata with AI-generated region captions.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AI-900 question from scratch — 985 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
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.