PartNeXt: Fine-Grained and Hierarchical 3D Part Annotations for Mesh Objects

A next-generation dataset for fine-grained and hierarchical 3D part understanding, released for the NeurIPS 2025 Dataset and Benchmark Track. It was created by researchers from ShanghaiTech University and is hosted on Hugging Face. The dataset page was last updated on October 26, 2025.

Use Cases

• Train 3D part segmentation models based on fine-grained part annotations.
• Develop hierarchical scene understanding algorithms based on the described part hierarchy.
• Benchmark next-generation 3D computer vision models on a dataset designed for the NeurIPS 2025 track.

Strengths

• Dataset is associated with a NeurIPS 2025 Dataset and Benchmark Track submission, indicating a research standard.
• Focuses on fine-grained and hierarchical part understanding, a specific and advanced task in 3D vision.
• Has an associated project page, paper, and toolkit, suggesting structured documentation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• License information is unknown, which could restrict commercial use.

Provenance

Source

ShanghaiTech University

Freshness

Last updated 2025-10-26 17:33:57; freshness should be verified.