2025 · Meta Reality Labs · Intermediate
Aria Gen 2 Pilot Dataset
Aria Gen 2 Pilot Dataset captures daily activities with research glasses and synchronized multimodal sensor data.
Direct answer
What does Aria Gen 2 Pilot Dataset contribute?
Aria Gen 2 Pilot Dataset captures daily activities with research glasses and synchronized multimodal sensor data.
Background
The dataset demonstrates Aria Gen 2's use for egocentric AI, contextual AI, robotics, and multimodal learning. It includes raw wearable sensor data and machine-perception outputs across activities like cleaning, cooking, eating, playing, and walking.
Problem
The work addresses a central constraint in Egocentric Data: building systems that learn useful representations or actions while remaining general enough to transfer beyond a single demonstration or environment.
Core idea
Aria Gen 2 Pilot Dataset captures daily activities with research glasses and synchronized multimodal sensor data.
Architecture and method
The dataset demonstrates Aria Gen 2's use for egocentric AI, contextual AI, robotics, and multimodal learning. It includes raw wearable sensor data and machine-perception outputs across activities like cleaning, cooking, eating, playing, and walking.
- Research-glasses egocentric capture
- Time-synchronized multimodal data
- Open tools through Project Aria
Results and impact
It connects smart-glasses hardware directly to the training-data layer needed for contextual assistants and robot imitation.
Prerequisites
- Wearable sensing
- Egocentric video
- Multimodal data
Recommended reading order
Read the explanation above, review the related topic pages, then use the primary-source links below to inspect the abstract, figures, experiments, and released implementation.
Primary sources
External links are provided after the context needed to evaluate the work.
Follow-up research
Related papers and concepts
Ego4D: Unscripted First-Person Video Dataset
Ego4D is a large first-person video dataset for studying what people see, remember, manipulate, and anticipate.
Ego-Exo4D: First- and Third-Person Skilled Activity Dataset
Ego-Exo4D pairs synchronized first-person and third-person video with language, gaze, audio, pose, and 3D signals.
V-JEPA 2: Self-Supervised Video Models for Physical Planning
V-JEPA 2 learns predictive video representations that support visual understanding and zero-shot robot control.
Open X-Embodiment and RT-X
Open X-Embodiment combines robot datasets across institutions and trains policies that transfer across embodiments.
Common questions
Frequently asked questions
What is the main idea of Aria Gen 2 Pilot Dataset?
Aria Gen 2 Pilot Dataset captures daily activities with research glasses and synchronized multimodal sensor data.
Why is Aria Gen 2 Pilot Dataset important?
It connects smart-glasses hardware directly to the training-data layer needed for contextual assistants and robot imitation.
What should I learn before reading Aria Gen 2 Pilot Dataset?
Recommended prerequisites are Wearable sensing, Egocentric video, Multimodal data.