UPLOAD IMAGE
Or drop a file

KEY FEATURES

Falcon enables you to detect inappropriate content in milliseconds and then you filter images based on your application requirements The API provides a list of labels with confidence scores to enable fine-grained control over images you want to allow. Different communities need different moderation rules. With Falcon you can easily set up rules that work best for your community or your brand. Via using the API result, you determine the suitability of content for your application. For example, images of partial nudity might be acceptable but images containing explicit nudity might not. To filter images, use the predicted labels array that is returned by the API.

FALCON API (V1.0)

Falcon offers an API service to integrate an accurate image content moderator into your applications and websites With the Falcon function, you can identify certain scenes in images that are specific to your business needs. The Falcon API designed to categorize images in different categories as below:
SEXUAL SCENE
Kissing, Sexually Provocative, Sexual Situation, Sexual Activity, Pornographic Scene, Adult Toys
EXPLICIT NUDITY
Bare Chested Male, Graphic Male Nudity, Breast, Graphic Female Nudity
PARTIAL NUDITY
Bikini or Lingerie, Revealing Clothes, Mini Skirt or Short, Male Swimwear or Underwear
WEAPONS
Gun, Missiles
VISUALLY DISTURBING
Dissected Body, Emaciated Body, Corpses, Hanging
VIOLENCE
Graphic Violence or Gore, Physical Violence, Self Injury
ALCOHOL
Drinking, Alcoholic Beverages
DRUG
Drug Products, Drug Use, Pills, Drug Paraphermalia
TOBACCO
Tobacco Products, Smoking
GAMBLING
Gambling
MONEY
Extremism

QUICK AND EASY TO SETUP

You only need a few lines of code to be up and running. We have worked hard to make it simple and quick to integrate with your product. You can set the confidence threshold that Falcon uses to detect inappropriate content by specifying the Min-Confidence input parameter. Labels are not returned for the content that is detected with lower confidence than Min-Confidence. Input: Image(Bytes) Min confidence
Output: JSON Response with detected labels and the prediction confidence

CONTENT MODERATION PYTHON EXAMPLE

import requests

with open('image.jpg', "rb") as img_file:
    source_bytes = img_file.read()

headers = {'Min-Confidence': '0.50'}
response = requests.post('https://asia-east2-falcon 293005.cloudfunctions.net/falcon', data=source_bytes, headers=headers)
import requests

url = 'https://site.com/image.jpg'

headers = {'Min-Confidence': '0.50'}
response = requests.post('https://asia-east2-falcon 293005.cloudfunctions.net/falcon', data=requests.get(url), headers=headers)
Specifying a value for Min-Confidence that is less than 50% is likely to return a high number of false- positive results. We recommend that you use a value that is less than 50% only when detection with a lower precision is acceptable. If you do not specify a value for Min-Confidence, Falcon API returns labels for inappropriate content which are detected with at least 70% confidence.

SAMPLE

The Moderation Label array contains labels in the preceding categories and estimated confidence in the accuracy of the recognized content. A top-level label is returned along with any second-level labels that were identified. For example, Falcon might return "Partial Nudity" with a high confidence score as a top-level label. That might be enough for your filtering needs. However, if it is necessary, you can use the confidence score of a second-level label (such as "Bikini or Lingerie") to obtain more granular filtering.
Falcon API Response: { "Version": "1.0", "Message": "succeed", "Code": 0, "Min-Confidence": 0.7, "ModerationLabels": { "Explicit Nudity": { "Confidence": 1, "Labels": { "Bare Chested Male": 1 } }, "partial Nudity": { "Confidence": 0.94, "Labels": { "Male Swimwear or Underwear": 0.94 } } } }
Image as input:
Labels are not returning the final moderation result. The confidence rates show how likely the image was containing each label's scene. Set and change your threshold at any time directly in your code. You can set the confidence threshold based on your business strategy and restrictions to detect inappropriate content and filter out them.

THEY TRUST US

Falcon content moderation technology have been integrated in multiple apps & websites and has scanned million of pictures