aiqbench analysis
This research quantifies the trade-offs between Unity simulation and controllable diffusion models for augmenting object detection datasets in data-scarce environments. The study finds that hybrid approaches combining real data with limited synthetic augmentation significantly outperform pure synthetic or pure real-data baselines, mitigating domain drift while enhancing generalization. For practitioners deploying computer vision in hazardous fields like construction, these findings offer a concrete strategy to optimize model robustness without relying solely on expensive physical data collection.
From Unity Simulation to Diffusion-Based Augmentation: Quantifying Dataset Balance for Robust Object Detection
arXiv:2609.38010v1 Announce Type: cross Abstract: Modern computer vision models achieve high accuracy when trained on large-scale annotated datasets. In critical domains such as construction safety monitoring, data collection is costly, hazardous, and ethically constrained. This paper presents a systematic study comparing two complementary data generation paradigms, (1) Unity Simulation-based rendering and (2) Controllable Diffusion-based generation (CIA), for object detection under real data-scarce conditions. A unified experimental framework enables controlled dataset mixing across real, simulated, and generative sources, while maintaining identical model and training settings. Quantitative evaluation using Precision, Recall, mAP, and custom $\Delta$-metrics, reveals that neither simulation nor generative augmentation alone achieves optimal transferability. Unity-only training yields an mAP@0.5 drop of $-50\%$ relative to real data, while CIA-only training shows a milder $-16.5\%$ degradation. Hybrid compositions significantly improve performance, with the 90\% real + 10\% Unity configuration achieving the best overall mAP@0.5 of $62.68\%$ ($+7.64\%$ over baseline), and the 90\% real + 10\% CIA configuration maximizing precision at $74.45\%$. Results demonstrate that limited synthetic inclusion enhances generalization, while excessive substitution induces domain drift.
Abstract:Modern computer vision models achieve high accuracy when trained on large-scale annotated datasets. In critical domains such as construction safety monitoring, data collection is costly, hazardous, and ethically constrained. This paper presents a systematic study comparing two complementary data generation paradigms, (1) Unity Simulation-based rendering and (2) Controllable Diffusion-based generation (CIA), for object detection under real data-scarce conditions. A unified experimental framework enables controlled dataset mixing across real, simulated, and generative sources, while maintaining identical model and training settings. Quantitative evaluation using Precision, Recall, mAP, and custom $\Delta$-metrics, reveals that neither simulation nor generative augmentation alone achieves optimal transferability. Unity-only training yields an mAP@0.5 drop of $-50\%$ relative to real data, while CIA-only training shows a milder $-16.5\%$ degradation. Hybrid compositions significantly improve performance, with the 90\% real + 10\% Unity configuration achieving the best overall mAP@0.5 of $62.68\%$ ($+7.64\%$ over baseline), and the 90\% real + 10\% CIA configuration maximizing precision at $74.45\%$. Results demonstrate that limited synthetic inclusion enhances generalization, while excessive substitution induces domain drift.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
I.4.8; I.2.10; I.2.6; I.6.7
arXiv:2609.38010 [cs.CV]
(or arXiv:2609.38010v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.38010 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
https://doi.org/10.1145/3787256.3787261 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 282 words
Related coverage
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · Sep 4, 2026
Topology-Informed Visual Prompting For Vision Language Action Policies
arXiv · Sep 22, 2026
EAGT: Echocardiography Augmentation for Generalisability and Transferability
arXiv · Jul 29, 2026