Computer Vision: Images, Faces & OCR

Question 1

An insurance company wants to automatically process car accident photos to identify if the image contains a car, a truck, or a motorcycle. Which computer vision capability is this?

Accepted Answer

Image Classification

Answer

Object Detection

Answer

Optical Character Recognition (OCR)

Answer

Facial Analysis

Question 2

You need to identify the exact location (bounding box coordinates) of pedestrians in a traffic camera image. Which service or feature should you use?

Accepted Answer

Object Detection

Answer

Image Classification

Answer

Image Tagging

Answer

Semantic Segmentation

Question 3

You want to build an application that notifies users if a photo they attempt to upload contains violent or adult content. Which Azure AI Vision feature should you use?

Accepted Answer

Inappropriate content detection (Adult/Racy content)

Answer

Face detection

Answer

Smart Cropping

Answer

OCR

Question 4

A retail company wants to train a model to detect their specific logo in social media photos. The logo is not recognized by standard models. Which service should they use?

Accepted Answer

Azure AI Custom Vision

Answer

Azure AI Vision

Answer

Azure AI Face

Answer

Computer Vision Read API

Question 5

You are using the Azure AI Vision Read API. What type of content can it extract?

Accepted Answer

Printed and handwritten text from images and PDF documents

Answer

Only printed text from images

Answer

Only numbers from invoices

Answer

Objects and faces

Question 6

You want to automatically generate smart thumbnails that keep the main subject of the photo centered, even if the aspect ratio changes. Which feature do you use?

Accepted Answer

Smart Cropping

Answer

Image Tagging

Answer

Object Detection

Answer

Auto-rotation

Question 7

In Azure AI Face, what is the difference between 'Detection' and 'Identification'?

Accepted Answer

Detection finds where a face is; Identification determines whose face it is.

Answer

They are the same.

Answer

Detection is for humans; Identification is for animals.

Answer

Detection requires a database; Identification does not.

Question 8

An application needs to allow users to log in by taking a selfie and comparing it to their stored profile photo. Which Face API operation is this?

Accepted Answer

Verification

Answer

Identification

Answer

Grouping

Answer

Detection

Question 9

Which service would you use to analyze a video file and get a transcription of the spoken audio, as well as identify when famous people appear?

Accepted Answer

Azure AI Video Indexer

Answer

Azure AI Vision

Answer

Azure AI Speech

Answer

Azure Media Services

Question 10

You are training an object detection model in Custom Vision. What metrics are used to evaluate its performance?

Accepted Answer

Precision, Recall, and mAP (Mean Average Precision)

Answer

RMSE and MAE

Answer

Confusion Matrix only

Answer

Silhouette Score