OakInk: 3D Hand-Object Interaction Data with MANO Pose Models

OakInk provides 3D hand-object interaction (HOI) data using the MANO model, released by the OakInk organization in 2022. It serves as a repository for 3D hand pose estimation and motion generation tasks involving human-object manipulation.

Use Cases

• 3D hand pose estimation using MANO model parameters
• Motion generation for hand-object interaction sequences
• Interaction transfer between objects with similar functional affordances

Strengths

• Peer-reviewed at CVPR 2022
• Includes MANO model parameters for 3D hand representation
• Supports interaction-transfer across object geometries

Limitations

• Requires domain expertise in 3D computer vision and MANO parameters
• Lack of explicit column-level documentation in the summary

Provenance

Source

CVPR 2022 publication 'OakInk: A Large-scale Knowledge Repository for Understanding Hand-Object Interaction'

Collection Method

Annotated 3D hand-object interaction data

Freshness

Updated as of December 2024.