MilitantHitchhiker / NudeNet

Lightweight nudity detection

Looking for contributors/ maintainers for this repo: I have become busy with other stuff in the last years, still trying to maintain this repo as it is the current best OSS option for nudity detection, Looking for interested mainttainer, who can add/ work on more features for this repo (with my help of course)

NudeNet: lightweight Nudity detection

https://nudenet.notai.tech/ in-browser demo (the detector is run client side, i.e: in your browser, images are not sent to a server)

pip install --upgrade "nudenet>=3.4.2"

from nudenet import NudeDetector
detector = NudeDetector()
# the 320n model included with the package will be used

detector.detect('image.jpg') # Returns list of detections

detector.detect_batch(['image_1.jpg', 'image_2.jpg']) # Returns list of [list of detections]

• Python package example in colab

• detect and detect_batch accept file path(s), opencv image(s), image bytes(s), open(image_path, 'rb') (buffereader) objects

Available models

Model	resolution trained	based on	onnx link	pytorch link
320n	320x320	ultralytics yolov8n	link	link
640m	640x640	ultralytics yolov8m	link	link

# To use the 640m model, download the onnx file and pass the path to the model_path argument

detector = NudeDetector(model_path="downloaded_640m.onnx path", inference_resolution=640)

• 320n is the default model and is included in the nudenet python package by default

detection_example = [
 {'class': 'BELLY_EXPOSED',
  'score': 0.799403190612793,
  'box': [64, 182, 49, 51]},
 {'class': 'FACE_FEMALE',
  'score': 0.7881264686584473,
  'box': [82, 66, 36, 43]},
 ]

nude_detector.censor('image.jpg') # returns censored image output path

# optional censor(self, image_path, classes=[], output_path=None) classes and output_path can be passed

all_labels = [
    "FEMALE_GENITALIA_COVERED",
    "FACE_FEMALE",
    "BUTTOCKS_EXPOSED",
    "FEMALE_BREAST_EXPOSED",
    "FEMALE_GENITALIA_EXPOSED",
    "MALE_BREAST_EXPOSED",
    "ANUS_EXPOSED",
    "FEET_EXPOSED",
    "BELLY_COVERED",
    "FEET_COVERED",
    "ARMPITS_COVERED",
    "ARMPITS_EXPOSED",
    "FACE_MALE",
    "BELLY_EXPOSED",
    "MALE_GENITALIA_EXPOSED",
    "ANUS_COVERED",
    "FEMALE_BREAST_COVERED",
    "BUTTOCKS_COVERED",
]

Docker

docker run -it -p8080:8080 ghcr.io/notai-tech/nudenet:latest

curl -F f1=@"images.jpeg" "http://localhost:8080/infer"

{"prediction": [[{"class": "BELLY_EXPOSED", "score": 0.8511635065078735, "box": [71, 182, 31, 50]}, {"class": "FACE_FEMALE", "score": 0.8033977150917053, "box": [83, 69, 21, 37]}, {"class": "FEMALE_BREAST_EXPOSED", "score": 0.7963727712631226, "box": [85, 137, 24, 38]}, {"class": "FEMALE_BREAST_EXPOSED", "score": 0.7709134817123413, "box": [63, 136, 20, 37]}, {"class": "ARMPITS_EXPOSED", "score": 0.7005534172058105, "box": [60, 127, 10, 20]}, {"class": "FEMALE_GENITALIA_EXPOSED", "score": 0.6804671287536621, "box": [81, 241, 14, 24]}]], "success": true}⏎

Some interesting projects based on NudeNet

1 - by https://github.com/w-e-w, censor extension ps://github.com/notAI-tech/issues/131