Cloud Vision SafeSearch

Flags adult, violent, medical and racy imagery.

Made by GoogleImage and video analysis

Visit cloud.google.com

About Cloud Vision SafeSearch

SafeSearch is a detection feature of the Cloud Vision API, made by Google. The feature returns likelihoods for five categories: adult, spoof, medical, violence and racy.

The Vision API features list describes SafeSearch as providing likelihood ratings for explicit content categories. Likelihood ratings are expressed as six values, from UNKNOWN to VERY_LIKELY.

Drafted from docs.cloud.google.com · not yet confirmed by Google

How it works

SafeSearch is a feature of Google's Cloud Vision API. Send an image and it returns five likelihoods (very unlikely to very likely) for adult, spoof, medical, violence and racy. It's one call, cheap, and backed by the models Google uses for its own search.

It's a blunt instrument: five categories, no custom training, no video. Good as a first filter or as a second opinion next to another classifier.

  • From the company

Cloud Vision SafeSearch, from its website
From the company

Reviews

Be the first to review it

Anonymous
Would you use it again?

Signed as your field and level, never your name.

Know a team weighing Cloud Vision SafeSearch? Send them this page.