Portal
NSFW Classification API
Services > Content Moderation > NSFW Classification
The Webit NSFW Classification REST API lets you effortlessly search between millons of YouTube, Dailymotion and Vimeo videos with a single solution.
Introduction
The Webit NSFW Classification API lets you effortlessly search between millons of YouTube, Dailymotion and Vimeo videos with a single solution. And it includes a FREE plan to get started with.
API testing and usage
If you want to start using the NSFW Classification service or just want to test it a little, just press the buttons below.
You will need a RapidAPI account for testing production endpoints. However, the whole process is completely FREE of charge and no credit card is required.
Below is an example of a not-easy-to-detect nudity. Neural Machine performs very well and understands the image contains nudity, despite of the confusing perspective.
{
    "status": "fail",
    "data": null,
    "message": "Failed to perform NSFW classification for the provided input. Your request has not been debited. Please check https:\/\/webit.ai\/status for additional troubleshooting."
}
Here is an example of a black and white image containing nudity. Neural Machine understands what's depicted in the image despite of skin tone colors.
{
    "status": "fail",
    "data": null,
    "message": "Failed to perform NSFW classification for the provided input. Your request has not been debited. Please check https:\/\/webit.ai\/status for additional troubleshooting."
}
Below there is an example of a picture depicting what may trigger a false positive nudity. Despite that, Neural Machine performs perfectly fine and identifies a totally safe image with "neutral" rating.
{
    "status": "fail",
    "data": null,
    "message": "Failed to perform NSFW classification for the provided input. Your request has not been debited. Please check https:\/\/webit.ai\/status for additional troubleshooting."
}
Below is an image of a shirtless person. It is marked as safe, but score variance denotes that the subject is not completely covered. A "moderate" rating is present. Such scores and ratings can be used to implement customized solutions based on thresholds and flags.