A mobile-friendly visual content moderation model, based on the work of F. C. Akyon obtained by fine-tuning EfficientNet-b0 on nsfw images to detect nudity and sexual content in images or video frames with high accuracy.

For a Demo, see: [viddexa/moderators]

Performance Comparison

Nsfw image detection performance of nsfw-detector-nano compared with Azure Content Safety AI and Falconsai nsfw image detection model.
F_safe and F_nsfw below are class-wise F1 scores for safe and nsfw classes, respectively.
Results show that nsfw-detector-nano performs better than Falconsai and is comparable to Azure AI with less number of parameters.

Model	F_safe	F_nsfw	Params
nsfw-detector-nano	96.91%	96.87%	4M
nsfw-detector-mini	97.90%	97.89%	17M
Azure AI	96.79%	96.57%	N/A
Falconsai	89.52%	89.32%	85M

Usage with moderators library

Install with pip install moderators then run:

python

from moderators import AutoModerator

model = AutoModerator.from_pretrained("viddexa/nsfw-detection-nano")
results = model("<path-to-image-file>")

probs = {k: v for r in results for k, v in r.classifications.items()}
predicted_label = max(probs, key=probs.get) 

Usage with transformers library

Install with pip install transformers then run:

python

from transformers import (
    AutoImageProcessor,
    AutoModelForImageClassification)
from PIL import Image
import torch

img = Image.open("<path-to-image-file>")

processor = AutoImageProcessor.from_pretrained("viddexa/nsfw-detection-nano", use_fast = False)
model=  AutoModelForImageClassification.from_pretrained("viddexa/nsfw-detection-nano")

with torch.no_grad():
    inputs = processor(images=img, return_tensors="pt")
    outputs = model(**inputs)
    logits = outputs.logits
    probs = torch.softmax(logits, dim=-1)
    pred_id = int(probs.argmax())
    print(model.config.id2label[pred_id])

Model Details

A binary nudity and sexual content classification model that classifies images either as Safe or NSFW. Model is obtained by fine-tuning google's Efficientnet-b0.

Developed by: [Kerem Bozgan, Abdullah Kırman, Fatih Çağatay Akyön]
Model type: [ConvNet]
License: [apache-2.0]
Finetuned from model: [google/efficientnet-b0]

Intended Use and Limitations

This model is designed to detect explicit nudity alone. An image containing suggestive nudity or risqué clothing is labeled as safe if there is no explicit nudity in the image.

Appropriate Use Cases:

Filtering explicit images in user-generated content platforms
Assisting moderation workflows for social apps and community forums
Parental control or restricted content filtering
Content review pipelines for games, and streaming services

In these settings, the model acts as an initial screening tool to help reduce moderator workload. A human review step is recommended for final decisions.

This model is not intended for:

Moral or ethical judgment of individuals or their bodies
Surveillance or monitoring of people in real-world environments
Determining sexual intent
Legal classification of pornography vs. art
Use in contexts affecting employment, policing, social credit, or identity

BibTeX entry and citation info

bibtex

@article{akyon2023nudity,
  title={State-of-the-art in nudity classification: A comparative analysis},
  author={Akyon, Fatih Cagatay and Temizel, Alptekin},
  booktitle={2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)},
  pages={1--5},
  year={2023},
  organization={IEEE}
}

Model Card Authors

Kerem Bozgan & Fatih Akyon