Docs / Moderation / Visual Moderation / Hate & Offensive Content Detection

Hate & Offensive Content Detection

offensive-2.0

Automatically detect images and videos depicting hateful content, such as Nazi symbols, offensive gestures, and supremacist imagery.

Replaces legacy model offensive.

Overview

The Offensive model is designed to detect images and videos of hateful content, such as Nazi, Confederate, supremacist, or terrorist symbols. It also identifies offensive gestures, such as middle fingers.

Nazi-era symbols: Photos and illustrations depicting Nazi symbols that are sometimes still used by neo-Nazi or supremacist groups
offensive.nazi
- Nazi hakenkreuz: Swastika symbol appropriated by the Nazi Party, appearing on objects from the Nazi era, costumes, drawings, cartoons, tattoos, etc.
- Nazi flag: Flags from the Nazi era (e.g., Nazi Party flag, Reichskriegsflagge, Reichsdienstflagge), found on flags, clothing, tattoos, and other objects
- Known Nazi dignitaries: Depictions or illustrations of prominent figures from the Nazi era
- Nazi salute: The arm-raised salute historically associated with the Nazi regime
- Other Nazi-era symbols such as the iron cross (military decoration of Prussia, later used in Nazi Germany, found on Nazi flags, costumes, medals, and other objects), the SS bolt (symbol derived from Schutzstaffel of Nazi Germany, symbolizing victory, typically present on Nazi-era objects), the Sonnenrad / sunwheel (ancient European symbol appropriated by the Nazis) or the Sturmabteilung (emblem of Hitler's SA troops)

Please note that swastikas can appear in non-hateful contexts, such as in Hinduism and Buddhism. In such cases, the swastika will be flagged in a separate `asian_swastika` class.

Asian swastika: Photos and illustrations showing swastikas unrelated to Nazi movements, instead associated with various cultures and religions
offensive.asian_swastika
- Asian swastika: Symbol with deep cultural and religious significance, such as in Hinduism and Buddhism, where it represents prosperity and good fortune
Supremacist symbols: Photos and illustrations depicting supremacist symbols or practices
offensive.supremacist
- KKK (Ku Klux Klan) members: Depictions of KKK members in uniform or during ceremonies
- Burning cross: Symbol of a practice associated with the KKK
- Blood drop cross: Primary symbol associated with the KKK
- Celtic cross / Odin's Cross: Square cross interlocked with a circle, commonly used by white supremacist groups
- Valknut / valknot: Old Norse symbol appropriated by some white supremacist groups, but historically significant in Viking culture
- Odal rune / othala rune: Symbol from the runic alphabet, used by certain supremacist groups promoting aryanism, but historically linked to Norse heritage
- Wolfsangel: Old runic symbol appropriated by groups promoting aryanism
Confederate symbols: Photos and illustrations depicting Confederate flags or their variations
offensive.confederate
- Confederate flag: Flag of the Confederate States of America, commonly found on flags, clothing, tattoos, and other objects
- Edited versions of the Confederate flag: Variations or modifications of the Confederate flag where the design is still clearly recognizable
Terrorist symbols: Photos and illustrations showing common terrorist flags
offensive.terrorist
- ISIS flag: IS / ISIL / ISIS / Daesh version of the Black Standard
Offensive gestures: Photos and illustrations depicting people giving the middle finger
offensive.middle_finger
- Individuals intentionally displaying the middle finger—an obscene hand gesture commonly used in Western culture as a sign of contempt—toward a person, object, or camera
The meaning of gestures can vary across cultures.

Related models

The following models can provide a useful complement to the offensive model:

Violence Detection: Detect physical violence and fights.
Military Detection: Detect military equipment and personnel.

Use the model (images)

If you haven't already, create an account to get your own API keys.

Detect hate & offensive content

Let's say you want to moderate the following image:

You can either share a URL to the image, or upload the image file.

Option 1: Send image URL

Here's how to proceed if you choose to share the image URL:


curl -X GET -G 'https://api.sightengine.com/1.0/check.json' \
    -d 'models=offensive-2.0' \
    -d 'api_user={api_user}&api_secret={api_secret}' \
    --data-urlencode 'url=https://sightengine.com/assets/img/examples/example-fac-1000.jpg'


# this example uses requests
import requests
import json

params = {
  'url': 'https://sightengine.com/assets/img/examples/example-fac-1000.jpg',
  'models': 'offensive-2.0',
  'api_user': '{api_user}',
  'api_secret': '{api_secret}'
}
r = requests.get('https://api.sightengine.com/1.0/check.json', params=params)

output = json.loads(r.text)


$params = array(
  'url' =>  'https://sightengine.com/assets/img/examples/example-fac-1000.jpg',
  'models' => 'offensive-2.0',
  'api_user' => '{api_user}',
  'api_secret' => '{api_secret}',
);

// this example uses cURL
$ch = curl_init('https://api.sightengine.com/1.0/check.json?'.http_build_query($params));
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
$response = curl_exec($ch);
curl_close($ch);

$output = json_decode($response, true);


// this example uses axios
const axios = require('axios');

axios.get('https://api.sightengine.com/1.0/check.json', {
  params: {
    'url': 'https://sightengine.com/assets/img/examples/example-fac-1000.jpg',
    'models': 'offensive-2.0',
    'api_user': '{api_user}',
    'api_secret': '{api_secret}',
  }
})
.then(function (response) {
  // on success: handle response
  console.log(response.data);
})
.catch(function (error) {
  // handle error
  if (error.response) console.log(error.response.data);
  else console.log(error.message);
});

See request parameter description

Parameter	Type	Description
url	string	URL of the image to analyze
models	string	comma-separated list of models to apply
api_user	string	your API user id
api_secret	string	your API secret

Option 2: Send image file

Here's how to proceed if you choose to upload the image file:


curl -X POST 'https://api.sightengine.com/1.0/check.json' \
    -F 'media=@/path/to/image.jpg' \
    -F 'models=offensive-2.0' \
    -F 'api_user={api_user}' \
    -F 'api_secret={api_secret}'


# this example uses requests
import requests
import json

params = {
  'models': 'offensive-2.0',
  'api_user': '{api_user}',
  'api_secret': '{api_secret}'
}
files = {'media': open('/path/to/image.jpg', 'rb')}
r = requests.post('https://api.sightengine.com/1.0/check.json', files=files, data=params)

output = json.loads(r.text)


$params = array(
  'media' => new CurlFile('/path/to/image.jpg'),
  'models' => 'offensive-2.0',
  'api_user' => '{api_user}',
  'api_secret' => '{api_secret}',
);

// this example uses cURL
$ch = curl_init('https://api.sightengine.com/1.0/check.json');
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, $params);
$response = curl_exec($ch);
curl_close($ch);

$output = json_decode($response, true);


// this example uses axios and form-data
const axios = require('axios');
const FormData = require('form-data');
const fs = require('fs');

data = new FormData();
data.append('media', fs.createReadStream('/path/to/image.jpg'));
data.append('models', 'offensive-2.0');
data.append('api_user', '{api_user}');
data.append('api_secret', '{api_secret}');

axios({
  method: 'post',
  url:'https://api.sightengine.com/1.0/check.json',
  data: data,
  headers: data.getHeaders()
})
.then(function (response) {
  // on success: handle response
  console.log(response.data);
})
.catch(function (error) {
  // handle error
  if (error.response) console.log(error.response.data);
  else console.log(error.message);
});

See request parameter description

Parameter	Type	Description
media	file	image to analyze
models	string	comma-separated list of models to apply
api_user	string	your API user id
api_secret	string	your API secret

API response

The API will then return a JSON response with the following structure:

                  
                  
{
  "status": "success",
  "request": {
      "id": "req_24DNGegGf1Mo0n4rpaRwZ",
      "timestamp": 1512898748.652,
      "operations": 1
  },
  "offensive": {
      "nazi": 0.01,
      "asian_swastika": 0.01,
      "confederate": 0.01,
      "supremacist": 0.01,
      "terrorist": 0.01,
      "middle_finger": 0.01
  },
  "media": {
      "id": "med_24DNJfN2BlCGPGQBoZ5dO",
      "uri": "https://sightengine.com/assets/img/examples/example-fac-1000.jpg"
  }
}

See response description

Successful Response

Status code: 200, Content-Type: application/json

Field	Type	Description
status	string	status of the request, either "success" or "failure"
request	object	information about the processed request
request.id	string	unique identifier of the request
request.timestamp	float	timestamp of the request in Unix time
request.operations	integer	number of operations consumed by the request
offensive	object	results for the model
media	object	information about the media analyzed
media.id	string	unique identifier of the media
media.uri	string	URI of the media analyzed: either the URL or the filename

Error

Status codes: 4xx and 5xx. See how error responses are structured.

Use the model (videos)

Detecting Hate & Offensive content in videos

Option 1: Short video

Here's how to proceed to analyze a short video (less than 1 minute):


curl -X POST 'https://api.sightengine.com/1.0/video/check-sync.json' \
  -F 'media=@/path/to/video.mp4' \
  -F 'models=offensive-2.0' \
  -F 'api_user={api_user}' \
  -F 'api_secret={api_secret}'


# this example uses requests
import requests
import json

params = {
  # specify the models you want to apply
  'models': 'offensive-2.0',
  'api_user': '{api_user}',
  'api_secret': '{api_secret}'
}
files = {'media': open('/path/to/video.mp4', 'rb')}
r = requests.post('https://api.sightengine.com/1.0/video/check-sync.json', files=files, data=params)

output = json.loads(r.text)


$params = array(
  'media' => new CurlFile('/path/to/video.mp4'),
  // specify the models you want to apply
  'models' => 'offensive-2.0',
  'api_user' => '{api_user}',
  'api_secret' => '{api_secret}',
);

// this example uses cURL
$ch = curl_init('https://api.sightengine.com/1.0/video/check-sync.json');
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, $params);
$response = curl_exec($ch);
curl_close($ch);

$output = json_decode($response, true);


// this example uses axios and form-data
const axios = require('axios');
const FormData = require('form-data');
const fs = require('fs');

data = new FormData();
data.append('media', fs.createReadStream('/path/to/video.mp4'));
// specify the models you want to apply
data.append('models', 'offensive-2.0');
data.append('api_user', '{api_user}');
data.append('api_secret', '{api_secret}');

axios({
  method: 'post',
  url:'https://api.sightengine.com/1.0/video/check-sync.json',
  data: data,
  headers: data.getHeaders()
})
.then(function (response) {
  // on success: handle response
  console.log(response.data);
})
.catch(function (error) {
  // handle error
  if (error.response) console.log(error.response.data);
  else console.log(error.message);
});

See request parameter description

Parameter	Type	Description
media	file	image to analyze
models	string	comma-separated list of models to apply
interval	float	frame interval in seconds, out of 0.5, 1, 2, 3, 4, 5 (optional)
api_user	string	your API user id
api_secret	string	your API secret

Option 2: Long video

Here's how to proceed to analyze a long video. Note that if the video file is very large, you might first need to upload it through the Upload API.


curl -X POST 'https://api.sightengine.com/1.0/video/check.json' \
  -F 'media=@/path/to/video.mp4' \
  -F 'models=offensive-2.0' \
  -F 'callback_url=https://yourcallback/path' \
  -F 'api_user={api_user}' \
  -F 'api_secret={api_secret}'


# this example uses requests
import requests
import json

params = {
  # specify the models you want to apply
  'models': 'offensive-2.0',
  # specify where you want to receive result callbacks
  'callback_url': 'https://yourcallback/path',
  'api_user': '{api_user}',
  'api_secret': '{api_secret}'
}
files = {'media': open('/path/to/video.mp4', 'rb')}
r = requests.post('https://api.sightengine.com/1.0/video/check.json', files=files, data=params)

output = json.loads(r.text)


$params = array(
  'media' => new CurlFile('/path/to/video.mp4'),
  // specify the models you want to apply
  'models' => 'offensive-2.0',
  // specify where you want to receive result callbacks
  'callback_url' => 'https://yourcallback/path',
  'api_user' => '{api_user}',
  'api_secret' => '{api_secret}',
);

// this example uses cURL
$ch = curl_init('https://api.sightengine.com/1.0/video/check.json');
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, $params);
$response = curl_exec($ch);
curl_close($ch);

$output = json_decode($response, true);


// this example uses axios and form-data
const axios = require('axios');
const FormData = require('form-data');
const fs = require('fs');

data = new FormData();
data.append('media', fs.createReadStream('/path/to/video.mp4'));
// specify the models you want to apply
data.append('models', 'offensive-2.0');
// specify where you want to receive result callbacks
data.append('callback_url', 'https://yourcallback/path');
data.append('api_user', '{api_user}');
data.append('api_secret', '{api_secret}');

axios({
  method: 'post',
  url:'https://api.sightengine.com/1.0/video/check.json',
  data: data,
  headers: data.getHeaders()
})
.then(function (response) {
  // on success: handle response
  console.log(response.data);
})
.catch(function (error) {
  // handle error
  if (error.response) console.log(error.response.data);
  else console.log(error.message);
});

See request parameter description

Parameter	Type	Description
media	file	image to analyze
callback_url	string	callback URL to receive moderation updates (optional)
models	string	comma-separated list of models to apply
interval	float	frame interval in seconds, out of 0.5, 1, 2, 3, 4, 5 (optional)
api_user	string	your API user id
api_secret	string	your API secret

Option 3: Live-stream

Here's how to proceed to analyze a live-stream:


curl -X GET -G 'https://api.sightengine.com/1.0/video/check.json' \
    --data-urlencode 'stream_url=https://domain.tld/path/video.m3u8' \
    -d 'models=offensive-2.0' \
    -d 'callback_url=https://your.callback.url/path' \
    -d 'api_user={api_user}' \
    -d 'api_secret={api_secret}'


# if you haven't already, install the SDK with 'pip install sightengine'
from sightengine.client import SightengineClient
client = SightengineClient('{api_user}','{api_secret}')
output = client.check('offensive-2.0').video('https://domain.tld/path/video.m3u8', 'https://your.callback.url/path')


// if you haven't already, install the SDK with 'composer require sightengine/client-php'
use \Sightengine\SightengineClient;
$client = new SightengineClient('{api_user}','{api_secret}');
$output = $client->check(['offensive-2.0'])->video('https://domain.tld/path/video.m3u8', 'https://your.callback.url/path');


// if you haven't already, install the SDK with 'npm install sightengine --save'
var sightengine = require('sightengine')('{api_user}', '{api_secret}');
sightengine.check(['offensive-2.0']).video('https://domain.tld/path/video.m3u8', 'https://your.callback.url/path').then(function(result) {
    // The API response (result)
}).catch(function(err) {
    // Handle error
});

See request parameter description

Parameter	Type	Description
stream_url	string	URL of the video stream
callback_url	string	callback URL to receive moderation updates (optional)
models	string	comma-separated list of models to apply
interval	float	frame interval in seconds, out of 0.5, 1, 2, 3, 4, 5 (optional)
api_user	string	your API user id
api_secret	string	your API secret

Moderation result

The Moderation result will be provided either directly in the request response (for sync calls, see below) or through the callback URL your provided (for async calls).

Here is the structure of the JSON response with moderation results for each analyzed frame under the data.frames array:

            
                  
{
  "status": "success",
  "request": {
    "id": "req_gmgHNy8oP6nvXYaJVLq9n",
    "timestamp": 1717159864.348989,
    "operations": 21
  },
  "data": {
    "frames": [
      {
        "info": {
          "id": "med_gmgHcUOwe41rWmqwPhVNU_1",
          "position": 0
        },
        "offensive": {
          "nazi": 0.01,
          "asian_swastika": 0.01,
          "confederate": 0.01,
          "supremacist": 0.01,
          "terrorist": 0.01,
          "middle_finger": 0.01
        },
      },
     ...
    ]
  },
  "media": {
    "id": "med_gmgHcUOwe41rWmqwPhVNU",
    "uri": "yourfile.mp4"
  },
}

You can use the classes under the offensive object to detect hateful or offensive content in the video.

Products

On this page

Hate & Offensive Content Detection

Overview

Related models

Use the model (images)

Detect hate & offensive content

Option 1: Send image URL

Option 2: Send image file

API response

Successful Response

Error

Use the model (videos)

Detecting Hate & Offensive content in videos

Option 1: Short video

Option 2: Long video

Option 3: Live-stream

Moderation result