Articulation-XL2.0: 11,300 3D Models for Automatic Rigging

Articulation-XL2.0 provides over 11,300 3D models specifically curated for automatic rigging and articulation tasks. Developed by researchers from Nanyang Technological University and Bytedance Seed for CVPR 2025, the collection includes a diverse-pose subset of 10.8k training and 500 testing samples.

Use Cases

• Training automatic rigging systems using the 10.8k training samples
• Benchmarking articulation-ready 3D models using the 500-sample test set
• Developing pose-aware 3D mesh deformation algorithms

Strengths

• 10.8k training samples in the diverse-pose subset
• Peer-reviewed research origin from CVPR 2025
• CC BY 4.0 open license for research accessibility

Limitations

• Specific file formats and metadata columns are not documented in the summary
• Narrow focus on articulation may exclude high-fidelity texture or material data

Provenance

Source

Nanyang Technological University, Bytedance Seed, and A*STAR (CVPR 2025)

Time Range

2025

Freshness

Updated in September 2025 following its CVPR 2025 publication.