DetectLewdImages Operator

Description

The DetectLewdImages operator detects inappropriate or lewd content in images using the Bumble Private Detector model. It analyzes images and returns a probability score indicating the likelihood of inappropriate content being present. The operator is designed to help moderate content by identifying potentially problematic images.

Model Information

  • Model: Bumble Private Detector

  • Source: Bumble, via HuggingFace Hub

  • Framework: TensorFlow

  • Input: Images (various formats supported)

  • Output: Probability score (0.0 to 1.0) indicating likelihood of inappropriate content

  • Usage: The model is used for content moderation and filtering inappropriate images from datasets or applications.

Dependencies

  • TensorFlow >= 2.19.0

  • HuggingFace Hub >= 0.30.2

How to Run the Tests

  1. Ensure that you are in the root directory of the feluda project.

  2. Install dependencies (in your virtual environment):

    uv pip install "./operators/detect_lewd_images"
uv pip install "feluda[dev]"

  3. Run the tests:

    pytest operators/detect_lewd_images/test.py

Usage

from feluda.factory import ImageFactory
from feluda.operators import DetectLewdImages

# Initialize the operator
operator = DetectLewdImages()

# Load an image
image = ImageFactory.make_from_url_to_path("https://example.com/image.jpg")

# Detect inappropriate content
probability = operator.run(image)

print(f"Probability of inappropriate content: {probability:.3f}")

# Cleanup
operator.cleanup()

Output Format

The operator returns a float value between 0.0 and 1.0:

  • 0.0: Very low probability of inappropriate content

  • 1.0: Very high probability of inappropriate content