moderation-all-resnext-2

An image moderation classifier model built on a ResNeXt architecture to identify 32 concepts, including GARM categories.

1
none
0.974
cigarettes-cigars-joints-blunts
0.025
vapes-e-cigs
0.008
drug-powders-and-paraphernalia
0.005
bongs-pipes-grinders-hookahs
0.003
smoking
0.001
drug-windowpanes-blotters
0.000
alcohol
0.000
cannabis-buds
0.000
cannabis-plant
0.000
cannabis-concentrates
0.000
pills-tablets-microdots
0.000
psilocybin-mushrooms-spores
0.000
sex-toys
0.000
self-harm
0.000
sexy-suggestive
0.000
male-genitals
0.000
nipples
0.000
bdsm-and-latex
0.000
bodysuit
0.000

Notes

moderation-all-resnext-2

Purpose

Image Moderation Classifier V2 model has been designed to moderate nudity, sexually-explicit, or otherwise harmful or abusive user-generated content (UGC) imagery. This is helpful to determine whether any given input image meets community and brand standards.

This model can identify 32 concepts in the following GARM Content Categories:

  • Adult & Explicit Sexual Content
  • Crime & Harmful acts to individuals and Society, Human Right Violations
  • Death, Injury or Military Conflict
  • Illegal Drugs / Tobacco / e-cigarettes / Vaping / Alcohol

Architecture

We based our image moderation classifier model on a ResNeXt architecture, and we trained it using a proprietary dataset gathered by our data strategy team. Given that ResNeXt is based on the residual ResNet approach, we have provided the following overview of both architectures:

ResNeXt

ResNeXt is a variation of the ResNet architecture that improves performance by increasing the width of the network. This is achieved by introducing a new dimension, cardinality C, which represents the size of the set of transformations. This approach is similar to the inception modules used in other architectures, allowing for more efficient processing and better performance.

ResNet

ResNet is a family of five ML CNN architectures that range from 18 to 152 layers, featuring one max pool and one average pool layer in each architecture, along with varying numbers of convolution layers.

Limitations

Due to the nature of our dataset, which heavily relies on real life imagery, the image moderation classifier model demonstrates weaker performance on hentai/cartoon harmful content.

Intended Use

Classification of inappropriate content which includes gore, drugs, explicit, suggestive, etc. in images and video. This model is intended to be used as a moderation tool to determine whether any given input image meets community and brand standards.

Resources

Image Moderation Classifier V2 Author

Clarifai - GitHub

  • ID
  • Model Type ID
    Visual Classifier
  • Input Type
    image
  • Output Type
    concepts
  • Description
    An image moderation classifier model built on a ResNeXt architecture to identify 32 concepts, including GARM categories.
  • Last Updated
    Feb 09, 2023
  • Privacy
    PUBLIC
  • Use Case
  • Toolkit
  • License
  • Share
  • Badge
    moderation-all-resnext-2