AI-102 Question 116
Select 2You are building an Azure AI-based application that uses the Computer Vision API to analyze images uploaded by users. When processing an image, the API returns the following JSON response:
{ "description": { "captions": [ { "text": "A group of people sitting at a table", "confidence": 0.95 } ] }, "tags": [ { "name": "people", "confidence": 0.98 }, { "name": "table", "confidence": 0.92 }, { "name": "indoor", "confidence": 0.88 } ], "faces": [ { "age": 34, "gender": "male" }, { "age": 28, "gender": "female" } ] }
Based on this response, which two statements about the analyzed image are correct?
- A
The image likely contains a group of people seated at a table.
- B
The API detected three faces in the image: one male and two females.
- C
The API identified the presence of tags such as 'people', 'table', and 'indoor' with high confidence.
- D
The description provided by the API has a confidence score of 0.88.
- E
The API is unable to determine the gender or age of individuals in the image.
Show answer and explanation
Correct answers: A, C
Explanation
The Azure Computer Vision API analyzes the image and generates a JSON response containing descriptions, tags, and facial attribute data. Accurate interpretation of this response is crucial for implementing AI solutions. The description and tags in the response indicate that the image likely contains a group of people at a table and includes relevant tags with high confidence scores. However, the response clearly shows only two detected faces with provided age and gender attributes, contrary to one of the incorrect options.
- A. Correct.
Correct: The description 'A group of people sitting at a table' is provided with a high confidence score of 0.95, indicating that this is likely accurate.
- B. Incorrect.
Incorrect: The response indicates the detection of only two faces, one male and one female, not three faces.
- C. Correct.
Correct: The tags 'people', 'table', and 'indoor' are identified with confidence scores of 0.98, 0.92, and 0.88 respectively, which are all relatively high.
- D. Incorrect.
Incorrect: The description confidence score is 0.95, not 0.88. This value is falsely interpreted here.
- E. Incorrect.
Incorrect: The API provides both age and gender information for the detected faces, as shown in the response.