3D-ARM-Gaze: 3D Arm Reaching Movements with Gaze in VR

3D-ARM-Gaze is a public dataset of natural arm movements with visual and gaze information collected in virtual reality. Bianca Lento and colleagues at the Centre National de la Recherche Scientifique published it in 2024. The dataset captures reaching movements from a precisely controlled, seated posture across a wide reachable space.

Use Cases

• Modeling natural arm reaching kinematics based on 3D movement data.
• Studying eye-hand coordination based on synchronized gaze and movement recordings.
• Developing VR interaction techniques based on natural movement patterns in a controlled seated posture.
• Benchmarking motion prediction algorithms based on precise 3D trajectory data.

Strengths

• Includes synchronized data streams for 3D arm movements and gaze information.
• Data collection was performed from a precisely controlled, comfortably seated posture.
• The repository provides accompanying Python analysis scripts and a Unity DataPlayer project for visualization.

Limitations

• Column-level documentation is absent; field semantics must be inferred from separate documentation files after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Centre National de la Recherche Scientifique

Collection Method

Recorded in a virtual reality environment during object-reaching tasks.