ModelNet10: 3D CAD Models Across 10 Object Categories

ModelNet10 contains 3D CAD models across 10 object categories, curated by SomTambe in May 2020. This version provides a compressed distribution of the standard Princeton ModelNet benchmark for 3D computer vision research.

Use Cases

• 3D object classification using the 10 category labels
• Point cloud sampling from the provided 3D mesh geometry
• Benchmarking 3D shape descriptors against the 10-class ground truth

Strengths

• 10 distinct object categories
• Standardized benchmark subset
• Compressed for reduced storage footprint

Limitations

• Unknown file format and encoding
• Lack of explicit documentation on preprocessing
• Temporal staleness

Provenance

Source

Princeton ModelNet

Collection Method

Curated subset of CAD models

Freshness

Last updated May 2020; no subsequent updates recorded.