arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.16770cs.CVcs.RO
收藏

具有方向可控性的3D全身抓取合成

3D Whole-body Grasp Synthesis with Directional Controllability

University of Amsterdam(阿姆斯特丹大学);Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所)

Georgios Paschalidis, Romana Wilschut, Dimitrije Antić, Omid Taheri, Dimitrios Tzionas

AI总结:

针对全身抓取合成中方向不合理、容器碰撞及搜索成本高的问题,本文提出CWGrasp,通过早期几何推理采样到达方向并支持左右手抓取,在GRAB和ReplicaGrasp上以更低成本超越基线。

摘要:

合成逼真抓取物体的3D全身对于动画、混合现实和机器人技术都很有用。这具有挑战性,因为双手和身体相对于彼此、被抓取物体以及局部场景(即支撑物体的容器)都需要看起来自然。此外,该任务的训练数据非常稀缺,而采集新数据成本高昂。近期工作通过分治方法超越了有限数据集;它首先生成一个“引导性”右手抓取,然后搜索与之匹配的身体。然而,引导手合成缺乏可控性和容器感知能力,因此很可能具有不合理的方向(即身体无法在不穿透容器的情况下与之匹配),需要通过大量后处理进行修正。此外,身体搜索需要穷举采样且成本高昂。这些都是严重局限。我们用一种名为CWGrasp的新方法解决这些问题。我们的核心思想是,“尽早”而非“太晚”执行基于几何的推理,可为推理提供丰富的“控制”信号。为此,CWGrasp首先从一个概率模型中采样合理的到达方向向量(随后用于手臂和手部),该模型通过从物体进行光线投射和碰撞检查构建。此外,CWGrasp独特地同时处理右手和左手抓取。我们在GRAB和ReplicaGrasp数据集上进行评估。CWGrasp以更低的运行时间和预算优于基线,同时所有组件都有助于性能。代码和模型可在https://gpaschalidis.github.io/cwgrasp获取。

英文摘要:

Synthesizing 3D whole bodies that realistically grasp objects is useful for animation, mixed reality, and robotics. This is challenging, because the hands and body need to look natural w.r.t. each other, the grasped object, as well as the local scene (i.e., a receptacle supporting the object). Moreover, training data for this task is really scarce, while capturing new data is expensive. Recent work goes beyond finite datasets via a divide-and-conquer approach; it first generates a "guiding" right-hand grasp, and then searches for bodies that match this. However, the guiding-hand synthesis lacks controllability and receptacle awareness, so it likely has an implausible direction (i.e., a body can't match this without penetrating the receptacle) and needs corrections through major post-processing. Moreover, the body search needs exhaustive sampling and is expensive. These are strong limitations. We tackle these with a novel method called CWGrasp. Our key idea is that performing geometry-based reasoning "early on," instead of "too late," provides rich "control" signals for inference. To this end, CWGrasp first samples a plausible reaching-direction vector (used later for both the arm and hand) from a probabilistic model built via ray-casting from the object and collision checking. Moreover, CWGrasp uniquely tackles both right and left-hand grasps. We evaluate on the GRAB and ReplicaGrasp datasets. CWGrasp outperforms baselines, at lower runtime and budget, while all components help performance. Code and models are available at https://gpaschalidis.github.io/cwgrasp.

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。

↑