AB
AiBoss
project

NSFW Detector - An open-source AI-powered tool for detecting inappropriate content, supporting the recognition of images, PDFs, and video files.

NSFW Detector (Not Safe For Work) is an open-source tool for detecting inappropriate content. NSFW Detector can identify inappropriate content in images, PDFs, and video files. The tool is based on Google's Vita-...

What is NSFW Detector?

NSFW Detector (Not Safe For Work, or NSFW for short) is an open-source tool for detecting inappropriate content. NSFW Detector can identify inappropriate content in images, PDFs, and video files. The tool is based on Google's...vit-base-patch16-224-in21kThe model supports CPU-only inference, requiring no GPU, and can run on most servers. NSFW Detector provides services based on an API, making it easy to integrate with other applications. It supports Docker deployment and is suitable for distributed deployment. NSFW Detector automatically accelerates inference using multiple CPUs, runs entirely locally to protect data security, and supports detection of various file types.

Main functions of NSFW Detector

  • Content detectionDetects whether files in images, PDFs, videos, and compressed packages contain inappropriate for working environments (NSFW) content.
  • CPU inferenceIt supports inference using the CPU, eliminating the need for GPU support and reducing hardware requirements.
  • Multi-CPU accelerationIt automatically uses multiple CPU cores to accelerate the inference process and improve processing speed.
  • Simple classificationThe content is divided into two categories, "nsfw" and "normal", to simplify the classification process.
  • API serviceProvides API interfaces for easy integration with other applications to achieve automated content detection.

Technical principles of NSFW Detector

  • Deep learning modelsUse Googlevit-base-patch16-224-in21kThe model is a deep learning model based on Vision Transformer (ViT) specifically designed for image recognition and classification tasks.
  • Feature extractionThe model learns to extract features from input images, videos, or documents based on training.
  • ClassifierThe extracted features are fed into a classifier, which determines whether the content belongs to the NSFW category based on the training data.

NSFW Detector project address

Application scenarios of NSFW Detector

  • social media platformsAutomatically detects and filters inappropriate content uploaded by users, protecting the community environment and complying with laws and regulations.
  • Content sharing websiteAutomatically detect when users upload images, videos, or documents to prevent the spread of inappropriate content.
  • Corporate intranetDeployed within the enterprise intranet to prevent employees from accessing or sharing inappropriate content and maintain a professional working environment.
  • Online advertising and marketingEnsure the compliance of advertising content to avoid damaging the brand image due to inappropriate content.
  • Online education platformProtect the learning environment and prevent students from accessing inappropriate learning materials.