A museum is building a mobile app feature that, when a visitor points their phone camera at an exhibit placard, must (1) produce a one-sentence natural-language description of what the photo shows and (2) pull out any printed text visible on the placard, such as the artifact's name and date. Which two AI capabilities does this feature require? (Select TWO.)
Select all correct answers, then click Submit.
Short Explanation
Read the requirement carefully: you need one capability that turns the photo into a natural-language sentence, and another that reads the printed characters off the placard. Image captioning gives you the sentence, while optical character recognition gives you the artifact's name and date. Single category labels or bounding boxes won't deliver either the sentence or the extracted text.
Full Explanation
The required capabilities are image captioning and optical character recognition. The feature has two distinct requirements: generating a one-sentence natural-language description of the photo, which is precisely what image captioning produces, and extracting the printed name and date from the placard, which is precisely what optical character recognition performs by reading text embedded in an image. Both capabilities are needed together because neither alone covers both requirements. The capability that draws a bounding box around each object instance is incorrect because it would locate and box individual objects in the frame, such as the placard itself or a nearby sculpture, but it does not generate a descriptive sentence and it does not read the words printed on the placard. The capability that assigns a single category label to an entire photo is incorrect because it would return a single category label like 'museum exhibit' or 'sculpture', which falls short of a full descriptive sentence and, like object detection, has no capability to extract printed text. Only the pairing of image captioning and optical character recognition satisfies both stated requirements of the museum's feature: a natural-language description and text extraction from the same photo.