Identity Verification

Virtual Try-on (VTON)

Virtual Try-on

Synthetic data for computer vision to enable more capable and ethical AI.

Technology to enable virtual try-on (VTON), or virtual fashion, has evolved at breakneck pace over the past several years. A new wave of digital-first, direct-to-consumer brands has emerged as VTON systems have become less expensive and more accessible to upstart businesses. Brick-and-mortar retailers, too, are enhancing their customers’ digital experiences with VTON for clothing, eyewear, jewelry, hats, and other accessories.

Body Type Variation

To build robust models and deliver engaging user experiences for a broad range of consumers, VTON developers require large amounts of body data spanning height, weight, and age. Synthesis Humans can be used to generate labeled datasets for training ML models on the entire range of human body types.
Body Type Variation
Pose Estimation​

Pose Estimation

Complete control over any possible human pose, all with accurate body landmarks to help developers build more robust pose estimation models for VTON. A large library of poses, actions and movements is mapped to thousands of identities and body types.

Makeup Application

Applying makeup virtually requires model accuracy for precise control over skin tones and textures. Different types of makeup, as well as makeup composition, often confuse VTON systems. With Synthesis Humans, you can model any possible combination of skin color and surfaces, including facial hair, imperfections, and other fine details.

Makeup Application
Clothing & Accessories

Clothing & Accessories

Building robust virtual try-on applications requires diverse datasets across clothing types and materials. Our system is able to produce millions of unique combinations of clothing spanning clothing type, materials, and textures, with advanced layering capabilities for complex styling.

Robust Labels

Pixel-perfect labels with accurate facial, body and hand landmarks to help developers build more robust VTON models. A large library of poses, actions and movements is mapped to thousands of identities and body types.
Robust Labels

Digital Human Datasets for Metaverse Applications

Synthesis AI is the technology leader for creating realistic digital humans. Our digital human datasets support the many different types of avatar characteristics, including skin tones, gestures, and facial expressions, needed for next generation virtual innovation in the Metaverse.

We focus on the fine details of human movement and characteristics, as well as human interactions with a given environment, to support the development of augmented and virtual reality applications, avatars and experiences in the Metaverse.

Robust Labels

Pixel-perfect labels with accurate facial, body and hand landmarks to help developers build more robust VTON models. A large library of poses, actions and movements is mapped to thousands of identities and body types.

Ready for a demo? We're here to help.

Explore datasets and labels with our Data Visualizer

X