AssemblyBench：面向复杂工业物体的物理感知装配
AssemblyBench: Physics-Aware Assembly of Complex Industrial Objects
提供了从多模态装配手册与 3D 零件几何联合预测装配顺序与 6-DoF 装配轨迹的设定与基准，可用于研究指令-几何对齐、部件级姿态估计与物理可行性约束下的运动生成。
原文依据 jointly predict assembly order and part assembly trajectories
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论从零件装配物体需要理解多模态指令、将其与三维部件关联，并为每个装配步骤预测物理上合理的六自由度运动。现有数据集聚焦于简化场景，忽略了工业装配中的形状复杂性与装配轨迹。我们提出 AssemblyBench，一个包含 2,789 个工业物体、多模态装配手册、对应三维零件模型以及零件装配轨迹的合成数据集。我们还提出一个基于 Transformer 的模型 AssemblyDyno，它利用装配手册与每个零件的三维形状，联合预测装配顺序与零件装配轨迹。AssemblyDyno 在装配姿态估计与轨迹可行性两方面均优于先前工作，其中轨迹可行性由我们基于物理的仿真进行评估。
英文原文摘要 840 字符
Assembling objects from parts requires understanding multimodal instructions, linking them to 3D components, and predicting physically plausible 6-DoF motions for each assembly step. Existing datasets focus on simplified scenarios, overlooking shape complexities and assembly trajectories in industrial assemblies. We introduce AssemblyBench, a synthetic dataset of 2,789 industrial objects with multimodal instruction manuals, corresponding 3D part models, and part assembly trajectories. We also propose a transformer-based model, AssemblyDyno, which uses the instructional manual and the 3D shape of each part to jointly predict assembly order and part assembly trajectories. AssemblyDyno outperforms prior works in both assembly pose estimation and trajectory feasibility, where the latter is evaluated by our physics-based simulations.