| --- |
| title: README |
| emoji: 📈 |
| colorFrom: indigo |
| colorTo: indigo |
| sdk: static |
| pinned: false |
| --- |
| |
| # FPV Labs |
|
|
| Open infrastructure for embodied AI. |
|
|
| Stera by FPV Labs turns the iPhone in your pocket into a spatial-data capture system, and gives you the open-source tooling to read, process, and export the result for training. |
|
|
| <p> |
| <a href="https://fpvlabs.ai"><img src="https://img.shields.io/badge/Website-black?style=for-the-badge" alt="Website"></a> |
| <a href="https://arxiv.org/abs/2605.05945"><img src="https://img.shields.io/badge/Paper-b31b1b?style=for-the-badge" alt="Paper"></a> |
| <a href="https://fpvlabs.ai/stera/docs"><img src="https://img.shields.io/badge/Docs-blue?style=for-the-badge" alt="Docs"></a> |
| <a href="https://github.com/fpv-labs"><img src="https://img.shields.io/badge/GitHub-181717?style=for-the-badge&logo=github&logoColor=white" alt="GitHub"></a> |
| <a href="https://fpvlabs.ai/app"><img src="https://img.shields.io/badge/iOS_App-999?style=for-the-badge&logo=apple&logoColor=white" alt="iOS App"></a> |
| <a href="https://fpvlabs.ai/discord"><img src="https://img.shields.io/badge/Discord-5865F2?style=for-the-badge&logo=discord&logoColor=white" alt="Discord"></a> |
| <a href="https://x.com/fpv_labs"><img src="https://img.shields.io/badge/X-000?style=for-the-badge&logo=x&logoColor=white" alt="X"></a> |
| </p> |
|
|
| ## Stera Suite |
|
|
| An end-to-end pipeline: Capture → Process → Evaluate → Export. |
|
|
| One iPhone Pro recording processed over stera gives you: |
| - Synchronized RGB + LiDAR depth |
| - 6-DoF pose (ARKit, globally anchored) |
| - 21-joint MANO hand poses (both hands) |
| - IMU & Env mesh |
| - Hierarchical action labels (session → sub-goal → episode → atomic span) |
|
|
| Exports to HDF5, MP4, PLY, MCAP, RRD, and LeRobot-compatible formats. |
|
|
| **Stera-10M** - 10M frames · 200 hours · 580 sessions · 108 min longest continuous capture · 70K+ atomic action spans · 20+ environments · all on consumer iPhones. |
|
|
| The first open dataset combining hour-plus continuous capture with depth, 6-DoF pose, and dense hand annotations on commodity hardware. |
|
|
|
|
