Templates

Overview

Synthesis AI offers a set of common scenarios and provides a JSON that programmatically creates images without the need for manual modification.

Workflow

The standard workflow for the Synthesis AI Face API is:

  1. Register Account, Download & Setup CLI
  2. Select Identities for the Face API with the Identities API
  3. Create Face API job with input JSON
  4. Download job output with CLI
  5. Parse outputs into your ML pipeline

Identities

Name Description Sample image

50 ids

50 human identities evenly distributed among gender,age and ethnicities.

100 ids

100 human identities evenly distributed among gender,age and ethnicities.

1000 ids

1000 human identities evenly distributed among gender,age and ethnicities.


Use Cases

Name Use case Description Sample image

Facial Landmark Estimation

Landmark Estimation

Camera focused on close-up images of the head. Variation in eye gaze, head turn, glasses, masks, facial hair, and background (HDRI)

Mobile Phone Re-ID

ID Verification

100 images (5 M, 5 F). Camera distance approximating a mobile phone camera. Variation includes eye gaze, head turn, glasses, masks, facial hair, and background (HDRI)

Teleconferencing Matting

Teleconferencing

Background matting is cutting out a foreground person from any and a key computer vision problem in the current era of online videoconferencing. Variation in glasses, headphones, facial hair, gestures and indoor background (HDRI). Emulating a "glow" from monitor using directional lights.

NIR rearview mirror in vehicle

Driver Monitoring

In-vehicle environment with a NIR camera located at the rearview mirror and foucsed on the driver. Variability in head pose, gaze, and accessories

RGB rearview mirror in vehicle

Driver Monitoring

In-vehicle environment with a RGB camera located at the rearview mirror and foucsed on the driver. Variability in head pose, gaze, and accessories

NIR dashboard in vehicle

Driver Monitoring

In-vehicle environment with a NIR camera located at the dashboard and focused on the driver. Variability in head pose, gaze, and accessories

RGB dashboard in vehicle

Driver Monitoring

In-vehicle environment with a RGB camera located at the dashboard and focused on the driver. Variability in head pose, gaze, and accessories

Static gestures

Body Pose Estimation

50 images (25M, 25F). Camera showing full body. Variation of 10 different gestures and background (HDRI)

Gesture animation

Body Pose Estimation

1 ID, 55 images each representing a frame of a gesture. Camera showing full body. Variation of background (HDRI).