BACK TO ALL BLOGS

Updates to Hive’s Best-in-Class Visual Moderation Model

Contents


Hive’s visual classifier is a cornerstone of our content moderation suite. Our visual moderation API has consistently been the best solution on the market for moderating key types of image-based content, and some of the world’s largest content platforms continue to trust Hive’s Visual Moderation model for effective automated enforcement on NSFW images, violence, hate, and more. 

As content moderation needs have evolved and grown, our visual classifier has also expanded to include 75 moderation classes across 31 different model heads. This is usually an iterative process – as our partners continue to send high volumes of content for analysis, we uncover ways to refine our classification schemes and identify new, useful types of content.

Recently, we’ve worked on broadening our visual model by defining new classes with input from our customers. And today, we’re shipping the general release of our latest visual moderation model, including three new classes to bolster our existing model capabilities:

  • Undressed to target racier suggestive images that may not be explicit enough to label NSFW
  • Gambling to capture betting in casinos or on games and sporting events
  • Confederate to capture imagery of the Confederate flag and graphics based on its design

All Hive customers can now upgrade to our new model to access predictions in these new classes at no additional cost. In this post, we’ll take a closer look at how these classes can be used and our process behind creating them.

New Visual Moderation Classes For Greater Content Understanding

Deep learning classifiers are most effective when given training data that illustrates a clear definition of what does and does not belong in the class. For this release, we used our distributed data labeling workforce – with over 5 million contributors – to efficiently source instructive labels on millions of training images relevant to our class definitions. 

Below, we’ll take a closer look at some visual examples to illustrate our ground truth definitions for each new class. 

Undressed

In previous versions, Hive’s visual classifier separated adult content into two umbrella classes: “NSFW,” which includes nudity and other explicit sexual classes, and “Suggestive,” which captures milder classes that might still be considered inappropriate. 

Our “Suggestive” class is a bit broad by design, and some customers have expressed interest in a simple way to identify the racier cases without also flagging more benign images (e.g., swimwear in beach photos). So, for this release, we trained a new class to refine this distinction: undressed

We wanted this class to capture images where a subject is clearly nude, even if their privates aren’t visible due to their pose, are temporarily covered by their hands or an object, or are occluded by digital overlays like emojis, scribbles, or shapes. To construct our training set, we added new annotations to existing training images for our NSFW and Suggestive classes and sourced additional targeted examples. Overall, this gave us a labeled set of 2.6M images to teach this ground truth to our new classifier. 

Here’s a mild example to help illustrate the difference between our undressed and NSFW definitions (you can find a full definition for undressed and other relevant classes in our documentation): 

Visual moderation model results for two versions of the Boticelli painting The Birth of Venus. Left: an unmodified version depicting (tasteful) female nudity, Right: an edited version with nudity covered by overlaid censor bars. Confidence scores in undressed are very high for both images, while the NSFW score drops significantly for the edited version.
Confidence scores for unedited version (left): undressed 1.00; general_nsfw 1.00; general_suggestive 0.00. Confidence scores for edited version (right): undressed 0.99; general_nsfw 0.35; general_suggestive 0.61

The first image showing explicit nudity is classified as both undressed and NSFW with maximum confidence. When we add a simple overlay over relevant parts of the image, however, the NSFW score drops far below threshold confidence while the undressed score remains very high. 

Platforms can use undressed to flag both nudity and more obviously suggestive images in a single class. For content policies where milder images are allowed but undressed-type images are not, we expect this class to significantly reduce any need for human moderator review to enforce this distinction. 

Gambling

Gambling was another type of content that frequently came up in customer feedback. This was a new undertaking for Hive, and building our ground truth and training set for this class was an interesting exercise in definitions and evaluating context in images. 

Technically, gambling involves a wager staked on an uncertain outcome with the intent of winning a prize. For practical purposes, though, we decided to consider evidence of betting as the key factor. Certain behavior – like playing a slot machine or buying a lottery ticket – is always gambling since it requires a bet. But cards, dice, and competitive games don’t necessarily involve betting. We found the most accurate approach to be requiring visible money, chips or other tokens in these cases in order to flag an image as gambling. Similarly, we don’t consider photos at races or sporting events to be gambling unless receipts from a betting exchange or website are also shown.  

To train our new class on this ground truth definition, we sourced and labeled a custom set of over 1.1M images. The new visual classifier can now distinguish between gambling activity and similar non-gambling behavior, even if the images are visually similar:

Visual moderation scores in the gambling class for four example images: people playing go fish normally (score 0.13; men playing poker with visible chips (score 1.00), yahtzee with dice and score sheets (score 0.07) and a close up of a craps table with chips and dice visible (score 1.00).

For more detailed information, you can see a full description of our gambling class here. Platforms that wish to moderate or identify gambling can access predictions from this model head by default after upgrading to this model release. 

Confederate Symbolism

Separately, many of our customers also expressed interest in more complete monitoring of visual hate and white nationalism, especially Confederate symbolism. For this release, we sourced and labeled over 1M images to train a new class for identifying imagery of the commonly used version of the Confederate flag. 

In addition to identifying photos of the flag itself, this new model head will also capture the Confederate “stars and bars” shown in graphics, tattoos, clothing, and the like. We also trained the model to ignore visually similar flags and historical variants that are not easily recognizable:

Confidence scores from the latest visual moderation model on four images related to the confederate flag: a combination with the "stars and stripes" (score: 0.99); a historical version (score: 0.00); a graphic (score: 0.99); and used noticeably in a tattoo (score: 0.97)

Along with our other hate classes, customers can now use predictions from our Confederate class to keep their online environments safe.

Improvements to Established Visual Moderation Classes

Beyond these new classes, we also focused on improving the model’s understanding around niche edge cases in existing model heads. For example, we leveraged active learning and additional training examples to address biases we occasionally found in our NSFW and Gun classifiers. This corrected some interesting biases where the model sometimes incorrectly identified studio microphones as guns, or mistook acne creams for other, less safe-for-work liquids. 

Final Thoughts

This release delivers our most comprehensive and capable Visual Moderation model yet to help platforms develop proactive, cost-effective protection for their online communities. As moderation needs become more sophisticated, we’ll continue to incorporate feedback from our partners and refine our content moderation models to keep up. Stay tuned for our next release with additional classes and improvements later this year.

If you have any questions about this release, please get in touch at support@thehive.ai or api@thehive.ai. You can also find a video tutorial for upgrading to the latest model configuration here. For more information on Visual Moderation more generally, feel free to reach out to sales@thehive.ai or check out our documentation