GPT-based Textile Pilling Classification Using 3D Point Cloud Data

TL;DR

This paper introduces TextileNet8, a new 3D point cloud dataset for textile pilling assessment, and proposes the PointGPT+NN model that achieves high accuracy in classifying pilling types.

Contribution

It presents the first publicly available 8-category 3D point cloud dataset for textile pilling and introduces a novel PointGPT+NN model for improved classification performance.

Findings

Achieved 91.8% overall accuracy on TextileNet8

Demonstrated the model's effectiveness on other datasets

First dataset of its kind for textile pilling assessment

Abstract

Textile pilling assessment is critical for textile quality control. We collect thousands of 3D point cloud images in the actual test environment of textiles and organize and label them as TextileNet8 dataset. To the best of our knowledge, it is the first publicly available eight-categories 3D point cloud dataset in the field of textile pilling assessment. Based on PointGPT, the GPT-like big model of point cloud analysis, we incorporate the global features of the input point cloud extracted from the non-parametric network into it, thus proposing the PointGPT+NN model. Using TextileNet8 as a benchmark, the experimental results show that the proposed PointGPT+NN model achieves an overall accuracy (OA) of 91.8% and a mean per-class accuracy (mAcc) of 92.2%. Test results on other publicly available datasets also validate the competitive performance of the proposed PointGPT+NN model. The proposed TextileNet8 dataset will be publicly available.