Domain Randomization for Object Detection in Manufacturing Applications using Synthetic Data: A Comprehensive Study

TL;DR

This study demonstrates that comprehensive domain randomization in synthetic data generation significantly improves object detection performance in manufacturing, achieving high accuracy with models trained solely on synthetic data.

Contribution

The paper introduces a detailed data generation pipeline and a new synthetic dataset, SIP15-OD, for manufacturing object detection, and provides extensive analysis of domain randomization effects.

Findings

Achieved high mAP scores using synthetic data alone.

Identified key factors like material properties and rendering methods.

Validated the effectiveness of domain randomization in sim-to-real transfer.

Abstract

This paper addresses key aspects of domain randomization in generating synthetic data for manufacturing object detection applications. To this end, we present a comprehensive data generation pipeline that reflects different factors: object characteristics, background, illumination, camera settings, and post-processing. We also introduce the Synthetic Industrial Parts Object Detection dataset (SIP15-OD) consisting of 15 objects from three industrial use cases under varying environments as a test bed for the study, while also employing an industrial dataset publicly available for robotic applications. In our experiments, we present more abundant results and insights into the feasibility as well as challenges of sim-to-real object detection. In particular, we identified material properties, rendering methods, post-processing, and distractors as important factors. Our method, leveraging these, achieves top performance on the public dataset with Yolov8 models trained exclusively on synthetic data; mAP@50 scores of 96.4% for the robotics dataset, and 94.1%, 99.5%, and 95.3% across three of the SIP15-OD use cases, respectively. The results showcase the effectiveness of the proposed domain randomization, potentially covering the distribution close to real data for the applications.