Skip to content

Analyze Videos

This section of the Microsoft AI-102: Designing and Implementing a Microsoft Azure AI Solution exam covers analyzing videos with Azure AI Video Indexer and Azure AI Vision Spatial Analysis. Below are study notes for each sub-topic, with links to Microsoft documentation, exam tips, and key facts

Use Azure AI Video Indexer to Extract Insights from a Video or Live Stream

📖 Docs: Azure AI Video Indexer overview

Overview

  • Video Indexer extracts insights from recorded or live video streams
  • Capabilities:
    • Speech-to-text transcription
    • Face detection and recognition
    • Object detection
    • Scene segmentation
    • Sentiment analysis from audio
    • OCR from video frames
    • Translation and subtitling

Key Points

  • Can process uploaded videos or connect to live streams
  • Integrates with Azure Media Services
  • Outputs JSON with structured insights for search and analytics

Use Case

Generating transcripts, detecting people and keywords in training videos, enabling video search

Use Azure AI Vision Spatial Analysis to Detect Presence and Movement of People in Video

📖 Docs: Azure Computer Vision - Spatial Analysis Tutorial

Overview

  • Spatial Analysis processes video feeds from cameras to understand movement and presence of people
  • Capabilities:
    • Person detection and counting
    • Line crossing detection
    • Dwell time analysis
    • Social distancing monitoring

Key Points

  • Runs on edge devices using containers
  • Requires GPU-enabled infrastructure
  • Privacy features such as anonymization (blurring faces/bodies) are included
  • Integrates with IoT and video analytics platforms

Exam Tip

Keywords like detecting people, movement, dwell time, or occupancy → Spatial Analysis

Quick‑fire revision sheet

  • 📌 Video Indexer = speech-to-text, OCR, sentiment, face/object detection, scene segmentation
  • 📌 Supports both uploaded video and live streams
  • 📌 Outputs structured insights for indexing and search
  • 📌 Spatial Analysis = detects presence, movement, dwell time of people
  • 📌 Runs in containers on edge devices, requires GPU
  • 📌 Includes privacy features (blurring, anonymization)