arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2004.07950cs.ROcs.AIcs.CVcs.LG
收藏

学习用于构建3D形状类别的视觉策略

Learning visual policies for building 3D shape categories

Inria(法国国家信息与自动化研究所(Inria))

Alexander Pashevich, Igor Kalevatykh, Ivan Laptev, Cordelia Schmid

摘要:

操纵和装配任务需要根据环境和最终目标进行非平凡的行动规划。以前在这个领域的工作通常是从已知的原语集合中组装特定的对象实例。相比之下,我们的目标是处理不同的基元集,并构建不同的对象的形状类别。给定一个类别的单个对象实例,例如拱门和二进制形状分类器,我们学习一个视觉策略来组装同一类别的其他实例。特别是,我们提出了一个拆卸过程,并学习状态策略,发现新的对象实例和他们的装配计划在状态空间。然后,我们在观察空间中渲染模拟状态,并学习热图表示来预测给定输入图像的替代动作。为了验证我们的方法,我们首先证明了它的效率,在状态空间中建立对象类别。然后,我们展示了我们的视觉政策的成功,从不同的基元建筑拱门。此外,我们证明了(i)我们的方法使用额外的原语重新组装对象的反应能力,以及(ii)我们的策略对于训练期间使用的类似构建块的不可见原语的鲁棒性能。我们的视觉装配策略是在没有真实图像的情况下训练的,在真实机器人上进行评估时,成功率高达95%。

英文摘要:

Manipulation and assembly tasks require non-trivial planning of actions depending on the environment and the final goal. Previous work in this domain often assembles particular instances of objects from known sets of primitives. In contrast, we aim to handle varying sets of primitives and to construct different objects of a shape category. Given a single object instance of a category, e.g. an arch, and a binary shape classifier, we learn a visual policy to assemble other instances of the same category. In particular, we propose a disassembly procedure and learn a state policy that discovers new object instances and their assembly plans in state space. We then render simulated states in the observation space and learn a heatmap representation to predict alternative actions from a given input image. To validate our approach, we first demonstrate its efficiency for building object categories in state space. We then show the success of our visual policies for building arches from different primitives. Moreover, we demonstrate (i) the reactive ability of our method to re-assemble objects using additional primitives and (ii) the robust performance of our policy for unseen primitives resembling building blocks used during training. Our visual assembly policies are trained with no real images and reach up to 95% success rate when evaluated on a real robot.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑