Vision Transformer Based Feature Extraction For Nsfw Embeddings

via “vision-transformer-feature-extraction”

image-classification model by undefined. 3,40,24,086 downloads.

Unique: Exposes full ViT architecture internals (patch embeddings, multi-head attention, layer-wise activations) rather than just final logits, enabling interpretable NSFW detection through attention map visualization and supporting transfer learning for custom content policies

vs others: Provides deeper model introspection than black-box APIs (Google Vision, AWS Rekognition), enabling researchers and platform teams to understand and customize NSFW boundaries rather than accepting fixed vendor definitions