What is the fastest, lowest-cost way to detect animals in user photos?

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?

  1. Send user-submitted images to the Cloud Vision API. Use object localization to identify all objects in the image and compare the results against a list of animals. Source Reference Answer
  2. Download an object detection model from TensorFlow Hub. Deploy the model to a Vertex AI endpoint. Send new user-submitted images to the model endpoint to classify whether each photo has an animal.
  3. Manually label previously submitted images with bounding boxes around any animals. Build an AutoML object detection model by using Vertex AI. Deploy the model to a Vertex AI endpoint Send new user-submitted images to your model endpoint to detect whether each photo has an animal.
  4. Manually label previously submitted images as having animals or not. Create an image dataset on Vertex AI. Train a classification model by using Vertex AutoML to distinguish the two classes. Deploy the model to a Vertex AI endpoint. Send new user-submitted images to your model endpoint to classify whether each photo has an animal.

Community Votes

A
100%

100% of anonymous learners picked answer A. Votes are pick records left by other test-takers — they are not the verified answer.

Community Insight

This question tests the trade-off between custom machine learning development and prebuilt Google Cloud APIs; the trap is choosing AutoML or a deployed model when the built-in Vision API is cheaper and faster to implement.

For a low-volume image upload use case requiring near real-time animal detection with minimal time and cost, the community strongly favors the Cloud Vision API over custom AutoML or TensorFlow model deployment. The Cloud Vision API provides a fully managed, prebuilt object localization capability that requires no custom training or labeling.

Selecting D (Vertex AutoML classification) because it is a simple binary classification task. This ignores the explicit priority on minimizing time and cost, as AutoML requires labeling, training, and endpoint deployment, whereas the Cloud Vision API works immediately.

Community Discussion (6 comments)

b1a8fae 👍 10 Selected: A
A. B would also work and I wonder if cost would be lower, but I think going with the google hosted service is most times the most likely choice to be correct.
rajshiv 👍 1 Selected: D
Option D is optimal as it uses image classification with Vertex AutoML, which is simple to implement, cost-effective, and scalable.
d6e1ae4 👍 1 Selected: D
The labeling process is simpler than object detection, as it's just a binary classification. AutoML simplifies the model creation process, reducing development time. For the relatively low volume of images (20 per day), this solution is likely to be cost-effective in the long run. Why not A? Cloud Vision is overkill for a binary classification and it is very expensive.
gscharly 👍 1 Selected: A
agree with b1a8fae
CHARLIE2108 👍 2 Selected: B
I went Option B
shadz10 👍 4 Selected: A
As minimising time and cost are of priority and considering the small subset of images I believe A is the best option

Comments & Corrections

No comments yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Submitted for moderation before publishing. Keep it helpful and respectful.

Expert Analysis

Why the Answer Is Correct

Option A uses the Cloud Vision API's object localization to identify all objects in the image, then filters results against a list of animals. This requires no custom training, no manual labeling, and no model deployment. The Cloud Vision API is a fully managed service, which minimizes both development effort and ongoing operational cost. For 20 photos per day, the pay-per-use pricing is highly economical and the API returns results in near real time.

The comment by the user with 10 likes states that "going with the google hosted service is most times the most likely choice to be correct," which aligns with the exam's emphasis on time and cost minimization. Another comment explicitly mentions that the small image subset makes the Cloud Vision API the best option.

Why the Other Options Are Wrong

Option B requires downloading a TensorFlow model and deploying it to a Vertex AI endpoint. Although feasible, this involves significant engineering effort for deployment, scaling, and monitoring, which contradicts the priority on minimizing development time. It also does not leverage a prebuilt service designed for generic object detection.

Option C requires manually labeling previously submitted images with bounding boxes and training an AutoML object detection model. This is overkill for a simple binary presence/absence question and requires a large, well-labeled dataset to achieve reliable accuracy—more time and cost than necessary.

Option D is the most common distractor because it uses binary image classification with AutoML, which is conceptually simpler than object detection. However, it still requires manual labeling, training, and endpoint deployment. As one comment notes, "Cloud Vision is overkill for a binary classification and it is very expensive," but that comment overlooks the fact that using AutoML also carries training and infrastructure costs. For a low-volume use case, the Cloud Vision API's per-request cost is very low and requires zero upfront ML expertise.

Community Comment Notes

Most community members voted for A, and the top comment points out that while B would also work, Google-hosted services are usually the safest exam choice. A few commenters chose D, but their reasoning typically underestimates the development overhead of AutoML. The commenter for B (2 likes) did not provide reasoning, and the A-voters emphasize the priority on "minimizing time and cost" as the deciding factor. The exam's suggested answer and majority consensus both verify A.

Official Reference

Exam Strategy

When the question stresses minimizing time and cost, look for a fully managed prebuilt API (like Cloud Vision) over custom AutoML or a deployed model. The only time you need AutoML is when no prebuilt API exists for the specific task or the API cannot achieve the required accuracy.

Related Analysis

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide