ClothesNet: An Information-Rich 3D Garment Model Repository with Simulated Clothes Environment

TL;DR

ClothesNet is a comprehensive 3D garment dataset with detailed annotations, enabling advanced computer vision and robotic tasks like classification, segmentation, and dressing in simulated and real environments.

Contribution

We introduce a large-scale, annotated 3D clothes dataset and establish benchmark tasks and simulated environments for robotic interaction with garments.

Findings

ClothesNet covers 4400 models across 11 categories.

Benchmark tasks demonstrate improved clothes perception accuracy.

Simulated environments facilitate robotic manipulation research.

Abstract

We present ClothesNet: a large-scale dataset of 3D clothes objects with information-rich annotations. Our dataset consists of around 4400 models covering 11 categories annotated with clothes features, boundary lines, and keypoints. ClothesNet can be used to facilitate a variety of computer vision and robot interaction tasks. Using our dataset, we establish benchmark tasks for clothes perception, including classification, boundary line segmentation, and keypoint detection, and develop simulated clothes environments for robotic interaction tasks, including rearranging, folding, hanging, and dressing. We also demonstrate the efficacy of our ClothesNet in real-world experiments. Supplemental materials and dataset are available on our project webpage.