发表机构
Technical University of Applied Sciences Augsburg; R. Waitz Data & Science（奥格斯堡应用技术大学; R. Waitz数据与科学公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究比较多种旋转表示预测SLM牙科零件构建方向，发现测试时增强是提升精度的关键，最佳表示依赖骨干网络。
AI 中文摘要
选择性激光熔化（SLM）制造牙科零件的构建方向通常由技术人员手动选择。我们将方向预测视为从技术人员标注的生产数据中对零件上轴进行监督机器学习，并测试哪种旋转表示能产生最佳结果。使用约2400个患者特异性牙科零件，我们在13种上轴表示上训练了ResNet-50多视图图像骨干和PointNeXt-S点云骨干，两者均经过预训练并端到端微调，这13种表示涵盖六种经典SO(3)参数化和七种直接在单位球面S^2上定义的表示。我们在测试集上报告预测上轴与真实上轴之间的测地角误差，分别在有和没有测试时增强（TTA）的情况下，TTA使用K=21个已知旋转。使用TTA时，八面体映射实现了最低的平均角误差（10.6°，ResNet-50）。总体而言，误差最低的三个结果是直接的S^2表示，尽管这可能反映了SO(3)目标中无监督面内分量的标签噪声，而非拓扑优势。von Mises-Fisher在使用PointNeXt-S训练时坍缩为接近恒定的预测，但使用ResNet-50时则不然。TTA在几乎所有表示和骨干上将平均角误差降低了31-73%。总体而言，对少量已知旋转的测试时增强是精度最一致的驱动因素，而最佳表示则强烈依赖于骨干网络。
英文摘要
Build orientation for selective laser melting (SLM) manufacturing of dental parts is usually chosen manually by technicians. We treat orientation prediction as supervised machine learning of the part's up-axis from technician-labeled production data, and test which rotation representations produce the best results. Using $n\approx2400$ patient-specific dental parts, we trained a ResNet-50 multi-view image backbone and a PointNeXt-S point-cloud backbone, both pretrained and fine-tuned end-to-end, on 13 up-axis representations spanning six classical $SO(3)$ parameterizations and seven representations defined directly on the unit sphere $S^2$. We report the geodesic angular error between predicted and ground-truth up-axis on a test set, with and without test-time augmentation (TTA) over $K=21$ known rotations. With TTA, the octahedral map achieves the lowest mean angular error ($10.6^\circ$, ResNet-50). The three lowest-error results overall are direct $S^2$ representations, though this may reflect label noise in the unsupervised in-plane component of the $SO(3)$ targets rather than a topological advantage. von Mises-Fisher collapses to a near-constant prediction when trained with PointNeXt-S but not with ResNet-50. TTA reduces mean angular error by 31-73 % across almost every representation and backbone. Overall, test-time augmentation over a small set of known rotations is the most consistent driver of accuracy, whereas the best-performing representation is strongly backbone-dependent.
