Skip to main content


DeepAR SDK allows developers to incorporate AR features into their applications. The SDK acts as an effect player for DeepAR effects with a host of built-in APIs that are needed to create robust AR solutions. DeepAR SDK is a base layer upon which all the DeepAR use-cases and solutions are built upon.

The best way to get started with DeepAR SDK is by visiting:

DeepAR SDK is available for:

How DeepAR works

DeepAR uses machine learning (ML) and deep learning models to track human features such as face position and facial expressions. Using that ML tracking data, DeepAR renders 2D and 3D graphics on top of camera feed creating an augmented reality experience. The process of analyzing the face and rendering graphics on top is done frame-by-frame. Application feeds camera frames into DeepAR, DeepAR does the ML analysis of the face, renders the graphics, and finally outputs the result to the screen.

DeepAR SDK consumes DeepAR effects that contain all the 3D assets, animations and logic that describes how the AR experience will look like.

DeepAR workflow