# Synthetic data for training and testing AI.

> Synthetic data & digital humans by Dream Awake Labs. Rendering pipelines that produce labeled, multi-view data of digital humans and physical scenes for training perception and reconstruction models, at a scale and precision real capture cannot reach.

The problem: Models that reconstruct a moving person from several cameras need many people, outfits, and motions seen by synchronized, calibrated cameras and labeled on every frame. Capturing that is slow and imprecise; rendering it is exact only if every camera sees the same instant.

Who it is for: For teams training perception, reconstruction, or physical-AI models that need labeled multi-view data real capture cannot deliver.

## A capture studio that exists only in the engine

For a team training multi-view human reconstruction models we built a rendering pipeline in Unreal Engine. Digital humans come from stock MetaHuman presets, varied in identity, outfit, and hair, and animated with SMPL-X motion capture retargeted to the MetaHuman skeleton. A tuning loop compares engine renders against the reference video and adjusts the conversion on its own.

A rig of calibrated color and infrared cameras renders every view on the same engine tick, so physics, hair, and particles match across views. Each camera writes depth, optical flow, normals, albedo, instance segmentation, object coordinates, 2D boxes, skeletal and deformed-mesh exports, and hair guides.

Batches run on a cloud render farm and land in a content-addressed registry as labeled multi-view clips. Visualization tooling checks the labels against a public 3D pose benchmark before a model trains on them.

- Pipeline diagram in six numbered stages: digital humans from stock engine presets; procedural variants of identity, outfit, and hair with retargeted body motion; a highlighted multi-camera rig rendering every view on one engine tick with calibrated color and infrared cameras; a cloud render farm; ground-truth passes for depth, optical flow, normals, albedo, segmentation, boxes, skeletons, and meshes; and a content-addressed asset registry. A tuning loop below compares renders with reference video. (From engine presets to labeled multi-view data. Drawn by us.): https://dreamawakelabs.com/work/synthetic-digital-humans/pipeline.svg

Case study: [Synthetic Data Pipelines for Digital Humans](https://dreamawakelabs.com/work/synthetic-digital-humans.md)

## How it works

- Unreal Engine 5 and MetaHuman: supply the digital humans, the physics, the hair, and the renderer, extended with our own plugins where the stock passes are not enough.
- SMPL-X retargeting: moves motion-capture datasets onto the engine skeleton, tuned automatically against the reference video.
- Movie Render Queue and a cloud render farm: render every camera on one engine tick and scale a batch to millions of frames.
- Ground-truth passes and a content-addressed registry: label every frame per camera and keep every asset and output traceable.

## What we can build

- **Digital human generation**: Identity, outfit, and hair variants generated from engine presets, animated with retargeted motion capture.
- **Motion retargeting**: SMPL-X and other mocap formats converted to engine skeletons, tuned automatically against reference video.
- **Synchronized multi-camera rendering**: Calibrated color and infrared cameras rendered on the same engine tick, with intrinsics and extrinsics in every output.
- **Ground truth**: Depth, optical flow, normals, albedo, segmentation, boxes, skeletons, and meshes per frame and per camera.
- **Render orchestration**: Cloud render farms, content-addressed asset registries, and batch runs that scale to millions of frames.
- **Dataset tooling**: Visualization and validation against public benchmarks, so labels are checked before a model trains on them.

## Projects

- [Synthetic Data Pipelines for Digital Humans](https://dreamawakelabs.com/work/synthetic-digital-humans.md): A rendering pipeline in Unreal Engine that produces labeled, multi-view video of digital humans for training models that reconstruct people from several cameras.

## A small labeled set for one rig

We render a small labeled multi-view dataset for one camera rig and your team checks it against its own model before anything scales.

What you get: The dataset, its calibration and label formats, and a written note on what to change before a full run.

What you bring:

- The camera rig: count, placement, intrinsics
- The labels and formats your training code reads
- A way to check the output against the model

- [Scope a synthetic data pipeline](https://dreamawakelabs.com/start?from=%2Fsynthetic-data)
- Email: hello@dreamawakelabs.com

## Other areas of expertise

- [Story & World Intelligence](https://dreamawakelabs.com/story-intelligence.md): Film search, AI characters, and interactive stories.
- [Interactive & Physical Experiences](https://dreamawakelabs.com/experiences.md): Experiences that respond to the people in the room.
- [AI agents & business automation](https://dreamawakelabs.com/agents.md): AI agents for everyday business work.
- [Multiplayer game development](https://dreamawakelabs.com/multiplayer.md): Multiplayer games and the servers behind them.
- [Agent-Ready Web Applications](https://dreamawakelabs.com/agent-ready-web.md): Web apps for people and AI agents.
- [Digital Twin Simulation](https://dreamawakelabs.com/digital-twins.md): Simulated people, spaces, and systems in a game engine.
- [Custom engineering](https://dreamawakelabs.com/software.md): Engineering for the parts no off-the-shelf product covers.