Object perception is fundamental for tasks such as robotic material handling and quality inspection. However, modern supervised deep-learning perception models require large datasets for robust automation under semi-uncontrolled conditions. The cost of acquiring and annotating such data for proprietary parts is a major barrier for widespread deployment. In this context, we release SynthRender, an open source framework for synthetic image generation with Guided Domain Randomization capabilities. Furthermore, we benchmark recent Reality-to-Simulation techniques for 3D asset creation from 2D images of real parts. Combined with Domain Randomization, these synthetic assets provide low-overhead, transferable data even for parts lacking 3D files. We also introduce IRIS, the Industrial Real-Sim Imagery Set, containing 32 categories with diverse textures, intra-class variation, strong inter-class similarities and about 20,000 labels. Ablations on multiple benchmarks outline guidelines for efficient data generation with SynthRender. Our method surpasses existing approaches, achieving 99.1% mAP@50 on a public robotics dataset, 98.3% mAP@50 on an automotive benchmark, and 95.3% mAP@50 on IRIS.

翻译：物体感知是机器人物料搬运与质量检测等任务的基础。然而，现代基于监督的深度学习感知模型需要大规模数据集，以在半非受控条件下实现鲁棒的自动化。针对专有部件获取并标注此类数据的成本，是广泛部署的主要障碍。在此背景下，我们发布了SynthRender，这是一个具备引导式域随机化能力的合成图像生成开源框架。此外，我们对近期从真实部件二维图像创建三维资产的现实到仿真技术进行了基准测试。结合域随机化，这些合成资产即便对于缺乏三维文件的部件，也能提供低开销、可迁移的数据。我们还引入了IRIS（工业虚实图像集），包含32个类别，具有多样化的纹理、类内差异、显著的类间相似性以及约20,000个标注。在多个基准测试上的消融实验，为使用SynthRender进行高效数据生成提供了指导原则。我们的方法超越了现有方案，在公开机器人数据集上达到99.1% mAP@50，在汽车基准测试上达到98.3% mAP@50，在IRIS上达到95.3% mAP@50。