nsfw-filter/nsfw-filter

A free, open source, and privacy-focused browser extension to block “not safe for work” content built using TensorFlow.js.

What it solves

It provides a privacy-focused way to block "not safe for work" (NSFW) content on the web. It prevents users from seeing inappropriate images by classifying them in real-time as they load on a page.

How it works

The project is a browser extension that uses TensorFlow.js to run AI models directly on the user's device. This ensures that no images are uploaded to a server and no data leaves the browser. Users can choose between two different classification models: a Vision Transformer (ViT-384) or a MobileNet model from NSFWJS.

Who it’s for

Web browser users who want to filter out NSFW content while maintaining their privacy.

Highlights

  • On-device classification using TensorFlow.js for maximum privacy.
  • Support for multiple AI models (ViT-384 and MobileNet).
  • Customizable filter strictness and handling options (blur, grayscale, or hide).
  • Ability to allow specific websites to bypass the filter.

Related

  • Project
  • Project
  • Project
  • Project
  • Project