Introduction
The Sensity Detection API runs a selection of automated authenticity tests on images and videos. The tests are designed to detect the latest AI-based media manipulation and synthesis techniques. Currently, the Sensity Detection API supports the following forms of media analysis:
Face Manipulation
Available for: images / videos
This test detects a class of deepfake manipulation that takes a real video of a person and alters or swaps their face with another person. The API can be called with input from files or URLs of videos and images online.
GAN Generated Faces
Available for: images
This test detects entirely synthetically generated images of faces, such as the photorealistic StyleGAN, as they may appear on social media profiles, accounts of financial services online, or dating apps.
Credentials
For access to a trial account, please contact us at support@sensity.ai
Getting Started
To get started with the API let’s consider an end-to-end example: analyzing a person’s headshot (let’s call it person.png) to assess if it was generated by a GAN model.
API Token
In order to start making automated API calls, we need to set up an API token for our already registered account.
This step requires login authentication first. You can authenticate your account with username and password in the Authorize button in the Swagger UI.
We can do so by making a POST request to /auth/tokens. You should now receive a JSON object containing a token field. Save this string securely, as this is your API credentials from now on.
To set the token in the Swagger UI, paste it in the Bearer field, after clicking the Authorize button.
When using the API programmatically, we recommend to export the token as an environment variable: export SENSITY_TOKEN=your_api_key
