Garment Particles: A 5D Point-Cloud Representation for Garment Generation and Editing

Garment Particles is a dataset representing garments as a 5D point-cloud that jointly encodes 2D sewing patterns and 3D geometry. The dataset supports the Garment Particles Flow (GPF) framework for generating garments from high-level inputs like text, images, and sketches. The dataset was created by georgeNakayama and was last updated on May 31, 2026.

Use Cases

• Generate 3D garment geometry based on textual descriptions using the text-to-garment capability mentioned.
• Create garment models from 2D sketches or images using the sketch/image-to-garment functionality described.
• Edit existing 3D garment designs through the various editing operations enabled by the representation.

Strengths

• The dataset is based on a novel 5D point-cloud representation that symmetrically encodes both 2D patterns and 3D geometry.
• It supports a framework for intuitive generation from multiple high-level inputs, as described in the paper.

Limitations

• Dataset size, row count, and file formats are unknown, limiting suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• License information is unknown, which may restrict usage.

Provenance

Source

georgeNakayama on Hugging Face

Collection Method

The method is described in the associated paper but not detailed in the provided input.

Freshness

Last updated 2026-05-31 23:44:19.