Search papers, labs, and topics across Lattice.
This paper reviews the industrial visual sim-to-real domain by categorizing it based on the availability of prior information, specifically distinguishing between CAD-available and CAD-unavailable settings. It highlights how the presence of explicit object geometry can enhance various processes such as rendering and calibration, while in CAD-unavailable scenarios, reliance shifts to alternative methods like feature distributions and vision-language priors. Empirical analysis using datasets like T-LESS/BOP and MVTec AD reveals that factors such as source-distribution design and detector capacity can significantly influence transfer success, challenging the notion that CAD render count alone is sufficient for effective deployment.
CAD availability dramatically alters the landscape of industrial visual sim-to-real, revealing that geometry isn't the only key to successful deployment.
Industrial visual sim-to-real is often described as transferring from synthetic images to real images, but industrial deployment usually involves a broader mismatch between available evidence and required decisions. A system may be built from CAD renderings, simulated RGB-D observations, normal reference images, synthetic defects, pretrained feature spaces, or language prompts, yet deployed under different sensors, lighting, materials, fixtures, calibration, production variation, and rare defect modes. This review reframes industrial visual sim-to-real as a domain-gap problem organized by prior availability. We distinguish CAD-available settings, where explicit object geometry can support rendering, calibration, pose estimation, segmentation, and test-time geometric verification; CAD-unavailable settings, where geometry is replaced by normal-reference appearance, feature distributions, teacher-student residuals, synthetic anomaly assumptions, foundation features, or vision-language priors; and boundary-prior settings, where approximate models, templates, reference views, or semantic correspondences preserve only part of the CAD role. This framing connects CAD-based detection and 6D pose-estimation literature with industrial anomaly and surface-inspection literature that is usually reviewed separately. To make the taxonomy concrete, we use empirical anchors on T-LESS/BOP, MVTec AD, and VisA. The anchors show that CAD render count alone does not close transfer; source-distribution design, detector capacity, and small real calibration can matter more. They also show that CAD at test time creates a distinct verification channel through mask, pose, and depth consistency, whereas CAD-unavailable inspection relies on calibrated normality and feature deviation. The review therefore argues against a single cross-task leaderboard and instead asks what prior grounds the deployment decision.
