EgoXtreme: Egocentric Object Pose Data Under Extreme Conditions

A dataset captured at 30 fps using Aria glasses, providing high-resolution 1408 x 1408 raw fisheye RGB images. It was created by author taegyoun88 for robust 6D object pose estimation in egocentric views under extreme environmental conditions. The dataset features 15 participants performing diverse interactions with 13 different objects.

Use Cases

• Training models for 6D object pose estimation based on high-resolution egocentric fisheye images.
• Benchmarking robustness of pose estimation algorithms under extreme environmental conditions mentioned in the description.
• Developing and evaluating models for human-object interaction analysis from a first-person perspective.

Strengths

• High-resolution 1408 x 1408 raw fisheye RGB images.
• Captured at 30 frames per second using Aria glasses.
• Includes data from 15 participants interacting with 13 different objects.

Limitations

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

Provenance

Source

huggingface

Collection Method

Captured using Aria glasses.

Freshness

Last updated 2026-04-06 04:18:22; freshness should be verified.