Guide-linked quiz · Medium Level
Visual Search in E-commerce Quiz
Test e-commerce visual search: image embeddings, candidate ranking, similarity versus identity, catalog quality, user-captured images, privacy, and
Question 1 of 8
A photo search returns a chair with a similar silhouette but different dimensions. What should the result communicate?
Preview every question8 questions
Answers and explanations appear as you take the quiz.
- A photo search returns a chair with a similar silhouette but different dimensions. What should the result communicate?
- That it is the identical chair
- That it is visually similar, while showing verified dimensions and item details
- That dimensions can be inferred from the photo
- That availability is guaranteed
- How does a common visual-search system retrieve candidates?
- Convert every image into a barcode
- Use only the product title text
- Match the shopper’s identity to a customer record
- Encode the query and catalog images as vectors and rank similar representations
- A retailer wants products found through Google Lens. What setup does Search Central describe?
- Product details in Merchant Center with product listings enabled and image practices followed
- Only a robots.txt file
- An AI-generated page title without product data
- An unindexed image that is blocked from Google
- Why do lighting and camera angle affect visual search?
- They change the product’s actual inventory
- They disable every image encoder
- They can change the visual features encoded from the query image
- They identify the shopper automatically
- Which evaluation separates exact-item retrieval from style matching?
- Combine all queries and measure clicks only
- Test separate query sets and judge the intent and attribute correctness of results
- Use only studio images
- Train and test on the same query photos
- What limitation can affect click-based relevance labels?
- Clicks depend on which products were exposed and where they appeared
- Clicks prove every result is a perfect match
- Exposure does not affect behavior
- Clicks always mean purchase satisfaction
- Why should a visual-search interface offer filters after retrieving candidates?
- Filters make images unnecessary
- Filters prove exact identity
- Users cannot refine visual queries
- Users may want to constrain details such as size, material, price, or availability
- What privacy practice is appropriate for uploaded query images?
- Keep every image indefinitely for unrelated training
- Publish query images with search results
- Provide clear notice and limit retention to the service’s stated need
- Use images to identify bystanders without notice
Keep testing yourself
More quizzes picked for your level.