SynSur: An end-to-end generative pipeline for synthetic industrial surface defect generation and detection

TL;DR

This paper introduces SynSur, a comprehensive pipeline for generating synthetic industrial defect data to enhance defect detection, demonstrating modest improvements when combined with real data across different domains.

Contribution

The paper presents a novel end-to-end diffusion-based pipeline for synthetic defect generation, annotation, and transfer, improving industrial defect detection datasets.

Findings

Synthetic defects can augment real data to improve detection performance.

The pipeline's effectiveness varies across different industrial inspection domains.

Synthetic-only training does not outperform real data but can complement it.

Abstract

The bottleneck in learning-based industrial defect detection is often limited not by model capacity, but by the scarcity of labeled defect data: defects are rare, annotations are expensive, and collecting balanced training sets is slow. We present an end-to-end pipeline for synthetic defect generation and annotation, combining Vision-Language-Model-based prompts, LoRA-adapted diffusion, mask-guided inpainting, and sample filtering with automatic label derivation, and demonstrates the potential of real data with realistic synthetic samples to overcome data scarcity. The evaluation is conducted on, a challenging dataset of pitting defects on ball screw drives, and then on a subset of the Mobile phone screen surface defect segmentation dataset (MSD) dataset to test cross-domain transfer. Beyond downstream detector performance, we analyze key stages of the pipeline, including prompt construction, LoRA selection, and sample filtering with DreamSim and CLIPScore, to understand which synthetic samples are both realistic and useful. Experiments with YOLOv26, YOLOX, and LW-DETR show that synthetic-only training does not replace real data. When combined with real data, synthetic defects can preserve performance and yield modest gains in selected BSData training regimes. The MSD transfer study shows that the overall pipeline structure carries over to a second industrial inspection domain, while also highlighting the importance of domain-specific adaptation and annotation-quality control. Overall, the paper provides an end-to-end assessment of diffusion-based industrial defect synthesis and shows that its strongest value lies in strengthening scarce real datasets rather than substituting for them.