Q: 4
You work for a pet food company that manages an online forum Customers upload photos of their
pets on the forum to share with others About 20 photos are uploaded daily You want to
automatically and in near real time detect whether each uploaded photo has an animal You want to
prioritize time and minimize cost of your application development and deployment What should you
do?
Options
Discussion
I keep going back and forth but leaning mostly A. Cloud Vision API does the job with almost zero setup, which is what they're asking for. If it was about custom accuracy or animal breeds, I'd reconsider. Not positive but seems safest for time/cost. Anyone see a case where D would truly be faster?
A , Cloud Vision API does the heavy lifting with almost no setup. No model training or data labeling needed, so you save loads of time and money. Pretty sure that's what they're looking for here.
I don’t think D is right. A is better since Cloud Vision API gives you object localization out of the box, no need to train or label datasets. It’s faster and cheaper for a simple animal/not animal check, pretty sure that’s what they want.
A is wrong, it'd only be right if you don't need breed or species. If the detection had to involve distinguishing animal types, object classification in B or D would be needed. For just animal/no animal and minimum dev, A's better though. Seen similar phrasing trip me up before.
Its A for me. Cloud Vision API lets you send the image right off and does the heavy lifting, so no model training or custom setup needed. Keeps it fast and cheap, which the question emphasizes. If they wanted breed/type detection or super rare animals, then maybe B or D. Open to pushback, not 100%.
Don't think D works best here, since it's more manual setup. A lowers dev time and cost, trap is overcomplicating.
Probably A. Had something like this in a mock last month, Cloud Vision API makes it basically plug-and-play for basic animal detection.
Maybe D here. I get that AutoML takes some setup but it lets you distinguish images without needing precise object locations, and it's still on Vertex AI, so simple enough. Could be missing something but D feels reasonable for a binary animal/no-animal check.
Honestly, D. This one seems closer to what the question is asking for.
I don’t think D is right. A sounds better here since Cloud Vision API does the detection with no training needed and saves a lot of time. D looks tempting if you missed the point about fast deployment, agree?
Be respectful. No spam.
