ScratchSim: A Procedural Synthetic Data Pipeline for Surface Scratch Detection

• URL: http://arxiv.org/abs/2607.27065v2
• Date: Thu, 30 Jul 2026 07:25:24 GMT
• Title: ScratchSim: A Procedural Synthetic Data Pipeline for Surface Scratch Detection
• Abstract summary: This paper presents a procedural rendering pipeline that generates large-scale annotated synthetic training data using Blender.<n>We evaluate four training strategies, namely synthetic-only, real-only, mixed, and fine-tuning from synthetic weights.<n>The proposed approach enables scalable defect detection without the burden of large real annotated datasets.
• Score: 7.759492039775529
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: While automated defect detection such as the detection of surface scratched is an important aspect in industrial quality control, the scarcity of annotated defect data make this task challenging. This paper presents a procedural rendering pipeline that generates large-scale annotated synthetic training data using BlenderProc, with configurable material appearance, camera modes, and domain randomization, producing automatic COCO-format annotations. To show the potential of our approach, we evaluate four training strategies, namely synthetic-only, real-only, mixed, and fine-tuning from synthetic weights, across two objects with different material properties and three lightweight edge-deployable detectors, YOLOX, YOLO26, and LW-DETR. Our evaluation show that fine-tuning from synthetic weights consistently outperforms real-only training, and that mixed training effectively recovers performance under scarce real-data conditions, with findings validated across both convolutional and transformer-based architectures. The proposed approach enables scalable defect detection without the burden of large real annotated datasets, making it practical for on-device industrial inspection. The pipeline scripts, 3D model, and both synthetic and real annotated scratch datasets for a glossy toy Ferrari car will be made available through the project website upon acceptance.

Related papers

• SynSur: An end-to-end generative pipeline for synthetic industrial surface defect generation and detection [4.219862887901752]
  We present an end-to-end pipeline for synthetic defect generation and annotation.<n>We analyze key stages of the pipeline, including prompt construction, LoRA selection, and sample filtering.<n>Overall, the paper provides an end-to-end assessment of diffusion-based industrial defect synthesis.
  arXiv  Detail & Related papers  (2026-04-29T12:57:32Z)
• Scaling Transformer-Based Novel View Synthesis Models with Token Disentanglement and Synthetic Data [53.040873127309766]
  We propose a token disentanglement process within the transformer architecture, enhancing feature separation and ensuring more effective learning.<n>Our method outperforms existing models on both in-dataset and cross-dataset evaluations.
  arXiv  Detail & Related papers  (2025-09-08T17:58:06Z)
• Synthetic Dataset Generation for Autonomous Mobile Robots Using 3D Gaussian Splatting for Vision Training [0.708987965338602]
  We propose a novel method for automatically generating annotated synthetic data in Unreal Engine.<n>We demonstrate that synthetic datasets can achieve performance comparable to that of real-world datasets.<n>This is the first application of synthetic data for training object detection algorithms in robot soccer.
  arXiv  Detail & Related papers  (2025-06-05T14:37:40Z)
• Bounding Box-Guided Diffusion for Synthesizing Industrial Images and Segmentation Map [50.21082069320818]
  We propose a novel diffusion-based pipeline for generating high-fidelity industrial datasets with minimal supervision.<n>Our approach conditions the diffusion model on enriched bounding box representations to produce precise segmentation masks.<n>Results demonstrate that diffusion-based synthesis can bridge the gap between artificial and real-world industrial data.
  arXiv  Detail & Related papers  (2025-05-06T15:21:36Z)
• Drive-1-to-3: Enriching Diffusion Priors for Novel View Synthesis of Real Vehicles [81.29018359825872]
  This paper consolidates a set of good practices to finetune large pretrained models for a real-world task.<n>Specifically, we develop several strategies to account for discrepancies between the synthetic data and real driving data.<n>Our insights lead to effective finetuning that results in a $68.8%$ reduction in FID for novel view synthesis over prior arts.
  arXiv  Detail & Related papers  (2024-12-19T03:39:13Z)
• SYNOSIS: Image synthesis pipeline for machine vision in metal surface inspection [1.1802456989915404]
  We introduce a complete pipeline which describes in detail how to approach image synthesis for surface inspection. The pipeline is in detail evaluated for milled and sandblasted aluminum surfaces.
  arXiv  Detail & Related papers  (2024-10-18T19:46:12Z)
• Hardness-Aware Scene Synthesis for Semi-Supervised 3D Object Detection [59.33188668341604]
  3D object detection serves as the fundamental task of autonomous driving perception. It is costly to obtain high-quality annotations for point cloud data. We propose a hardness-aware scene synthesis (HASS) method to generate adaptive synthetic scenes.
  arXiv  Detail & Related papers  (2024-05-27T17:59:23Z)
• PITA: Physics-Informed Trajectory Autoencoder [7.394156899576076]
  Generative models can be used to augment real-world datasets with generated data to produce edge case scenarios. We propose the Physics-Informed Trajectory Autoencoder (PITA) architecture, whichcorporates a physical dynamics model into the loss function of the autoencoder.
  arXiv  Detail & Related papers  (2024-03-18T12:37:41Z)
• CarPatch: A Synthetic Benchmark for Radiance Field Evaluation on Vehicle Components [77.33782775860028]
  We introduce CarPatch, a novel synthetic benchmark of vehicles. In addition to a set of images annotated with their intrinsic and extrinsic camera parameters, the corresponding depth maps and semantic segmentation masks have been generated for each view. Global and part-based metrics have been defined and used to evaluate, compare, and better characterize some state-of-the-art techniques.
  arXiv  Detail & Related papers  (2023-07-24T11:59:07Z)
• A New Benchmark: On the Utility of Synthetic Data with Blender for Bare Supervised Learning and Downstream Domain Adaptation [42.2398858786125]
  Deep learning in computer vision has achieved great success with the price of large-scale labeled training data. The uncontrollable data collection process produces non-IID training and test data, where undesired duplication may exist. To circumvent them, an alternative is to generate synthetic data via 3D rendering with domain randomization.
  arXiv  Detail & Related papers  (2023-03-16T09:03:52Z)
• Towards 3D Scene Understanding by Referring Synthetic Models [65.74211112607315]
  Methods typically alleviate on-extensive annotations on real scene scans. We explore how synthetic models rely on real scene categories of synthetic features to a unified feature space. Experiments show that our method achieves the average mAP of 46.08% on the ScanNet S3DIS dataset and 55.49% by learning datasets.
  arXiv  Detail & Related papers  (2022-03-20T13:06:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.