GAT-Grasp：面向任务感知机器人抓取的手势驱动可供性迁移
GAT-Grasp: Gesture-Driven Affordance Transfer for Task-Aware Robotic Grasping
AI总结：
本文提出GAT-Grasp框架，通过手势驱动的可供性迁移从人-物交互视频中提取抓取知识，实现对新物体和杂乱环境的零样本任务感知抓取。
摘要：
在多样化物体和环境中实现精确且可泛化的抓取，对于智能协作机器人系统至关重要。然而，现有方法往往在模糊的可供性推理和对未见物体的适应性方面存在不足，导致抓取执行效果欠佳。本文提出GAT-Grasp，一种手势驱动的抓取框架，直接利用人类手势引导生成具有合适位置和姿态的任务特定抓取位姿。具体而言，我们引入一种基于检索的可供性迁移范式，利用手势与物体可供性之间的隐式关联，从大规模人-物交互视频中提取抓取知识。通过消除对预先给定物体先验的依赖，GAT-Grasp实现了对新物体和杂乱环境的零样本泛化。真实世界评估证实了其在多样化和未见场景中的鲁棒性，展示了在复杂任务设置下可靠的抓取执行能力。
英文摘要：
Achieving precise and generalizable grasping across diverse objects and environments is essential for intelligent and collaborative robotic systems. However, existing approaches often struggle with ambiguous affordance reasoning and limited adaptability to unseen objects, leading to suboptimal grasp execution. In this work, we propose GAT-Grasp, a gesture-driven grasping framework that directly utilizes human hand gestures to guide the generation of task-specific grasp poses with appropriate positioning and orientation. Specifically, we introduce a retrieval-based affordance transfer paradigm, leveraging the implicit correlation between hand gestures and object affordances to extract grasping knowledge from large-scale human-object interaction videos. By eliminating the reliance on pre-given object priors, GAT-Grasp enables zero-shot generalization to novel objects and cluttered environments. Real-world evaluations confirm its robustness across diverse and unseen scenarios, demonstrating reliable grasp execution in complex task settings.
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。
