Bounding Box-Guided Diffusion for Synthesizing Industrial Images and Segmentation Map
This addresses the costly and time-consuming data acquisition for industrial defect segmentation, though it appears incremental as it builds on existing diffusion and layout-conditioned methods.
The paper tackles the problem of generating synthetic industrial datasets for defect segmentation by proposing a diffusion-based pipeline that conditions on bounding boxes to produce accurate segmentation masks, demonstrating improved defect consistency and spatial accuracy compared to existing methods.
Synthetic dataset generation in Computer Vision, particularly for industrial applications, is still underexplored. Industrial defect segmentation, for instance, requires highly accurate labels, yet acquiring such data is costly and time-consuming. To address this challenge, we propose a novel diffusion-based pipeline for generating high-fidelity industrial datasets with minimal supervision. Our approach conditions the diffusion model on enriched bounding box representations to produce precise segmentation masks, ensuring realistic and accurately localized defect synthesis. Compared to existing layout-conditioned generative methods, our approach improves defect consistency and spatial accuracy. We introduce two quantitative metrics to evaluate the effectiveness of our method and assess its impact on a downstream segmentation task trained on real and synthetic data. Our results demonstrate that diffusion-based synthesis can bridge the gap between artificial and real-world industrial data, fostering more reliable and cost-efficient segmentation models. The code is publicly available at https://github.com/covisionlab/diffusion_labeling.