# Repeated pattern detection on fabric: A survey and novel approach

## TL;DR

This paper surveys and introduces a new method for detecting repeating patterns on fabric images with high precision and minimal training data.

## Contribution

The novel method detects fabric patterns with sub-pixel accuracy and requires no training data, using automatic calibration with limited supervision.

## Key findings

- The proposed method achieves 96% recall and less than 0.5 pixel alignment error on a synthetic dataset.
- The method is competitive with existing baselines and suitable for real-world applications.
- A synthetic dataset and code for its generation are released to support future research.

## Abstract

Modern textile industries frequently apply patterns, such as brand logos or motifs, in near-regular arrangements to create visually appealing products. Consequently, the application of computer vision for pattern recognition is highly valuable for automating production chains and reducing waste. In this work, we address the challenging task of automatically detecting repeating patterns on fabric images, accounting for real-world complexities such as variable lighting and intentional pattern variance. We begin with an in-depth literature review on repeated pattern detection, highlighting current trends, organizing them into a hierarchy of sub-tasks, and discussing the novelty of each paper. Subsequently, we propose a novel method to solve our specific instance of this problem, focusing on detecting patterns with sub-pixel accuracy. We conduct extensive experiments to compare its performance against several baselines from the literature. Our method can be applied with high precision to real-world problems without requiring training data, instead using an automatic calibration procedure with limited human supervision. On a small synthetic dataset, our method detects repeated patterns with a 96% recall rate and an average alignment error of less than 0.5 pixels in just a few seconds, making it competitive with all tested baselines. Finally, we release our dataset and the code for its generation to encourage further research in this area.