Project Aria Tools

Project Aria Tools is a suite of C++/Python utilities to help researchers expand the horizons of Augmented Reality, Machine Perception and Artificial Intelligence with Project Aria. It is designed to make it easier to use Aria data and its open datasets. It supports both Aria Gen1 and Aria Gen2 data.

---

🚀 What's New in Aria Gen2

Aria Gen2 introduces significant hardware and software improvements with full API support in this 2.0.0 release.

Hardware & Sensors

• 12MP RGB camera, 4 CV cameras (wider FOV, HDR, front-facing stereo), 2 eye tracking cameras
• New sensors: Proximity, contact microphone, PPG health, ambient light, GNSS
• 6-8 hour battery life, foldable form factor, direct interactivity with open-air speakers

On-Device Machine Perception

On-device algorithms powered by a custom Meta co-processor:

• Eye Tracking, Hand Tracking (21 keypoints), VIO/SLAM (20Hz + 800Hz high-freq trajectory)

Software & Tools

• Unified APIs: Same Python/C++ interface for both Gen1 and Gen2 data
• New Tools: aria_rerun_viewer (interactive 3D visualization), gen2_mp_csv_exporter, upgraded vrs_health_check
• Enhanced Streaming: USB/wireless sensor streaming with on-device perception signals

---

📖 Documentation

Aria Gen2 Documentation - NEW! ✨

• Gen2 Documentation - Complete guide for Aria Gen2 data and tools
  • Research Tools APIs
  • Python/C++ examples
  • Data formats and specifications
  • On-device ML features

Aria Gen1 Documentation

• Gen1 Documentation - Legacy documentation for Aria Gen1

---

📚 Interactive Python Tutorials (Google Colab)

Aria Gen2 Tutorials - NEW! ✨

Comprehensive tutorials covering Aria Gen2 data processing:

1. VrsDataProvider Basics - Load an Aria VRS file and access its streams.

2. Device Calibration - Sensor intrinsics and extrinsics, projection, undistortion.

3. Sequential Multi-Sensor Access - The queued API for streaming several sensors in timestamp order.

4. Device Time Alignment - Time domains, timestamp queries, and multi-device alignment over SubGHz.

5. MPS Basics - MPS output layout, MpsDataPathsProvider and MpsDataProvider.

6. VIO and Trajectory - Device pose: on-device VIO, and the MPS trajectory and semi-dense point cloud.

7. Hand Tracking - Hands from both sources: the on-device stream and the MPS result.

8. Eye Tracking - Gaze from all three sources: on-device geometric, on-device ML, and MPS.

9. Neural Band sEMG - The Meta Neural Band emg stream, its batch model and its two clocks.

Aria Gen1 Tutorials

• Aria VRS Data Provider

• Reading and using Aria Machine Perception Services output (SLAM, Eye Tracking, Hand Tracking data)

---

🗂️ Open Datasets

Aria Gen2 Datasets

• Aria Gen2 Pilot Dataset: Dataset link
  • Multi-participant indoor and outdoor recordings
  • Full on-device outputs (eye tracking, hand tracking, VIO)
  • SubGHz-synchronized multi-device captures
  • High-quality MPS outputs (SLAM, point clouds, trajectories)

Aria Gen1 Datasets

• Aria Everyday Activities:
  • Dataset link
  • 
• Aria Digital Twin:
  • Dataset link
  • 
• Aria Synthetic Environments:
  • Dataset link

---

How to Contribute

We welcome contributions! Go to CONTRIBUTING and our CODE OF CONDUCT for how to get started.

License

Project Aria Tools are released by Meta under the Apache 2.0 license.