MapleGrasp：基于掩码引导的语言驱动高效机器人抓取
摘要
通过 natural language 命令对未知对象的机器人操作仍具有挑战性。语言驱动机器人抓取(LDRG)从 natural language 查询和 RGB-D 图像中预测稳定抓取姿态。我们提出 MapleGrasp，一种利用 mask-guided feature pooling 实现 efficient vision-language driven grasping 的 novel framework。Our two-stage training first predicts segmentation masks from CLIP-based vision-language features。The second stage pools features within these masks to generate pixel-level grasp predictions，improving efficiency，and reducing computation。Incorporating mask pooling results in a 7% improvement over prior approaches on the OCID-VLG benchmark。Furthermore，我们引入 RefGraspNet，一个 eight times larger than existing alternatives 的 open-source dataset，显著增强 model generalization for open-vocabulary grasping。MapleGrasp scores a strong grasping accuracy of 89% when compared with competing methods in the RefGraspNet benchmark。Our method achieves comparable performance to larger Vision-Language-Action models on the LIBERO benchmark，and shows significantly better generalization to unseen tasks。Real-world experiments on a Franka arm demonstrate 73% success rate with unseen objects，surpassing competitive baselines by 11%。Code is provided in our github repository。
引用
@article{arxiv.2506.06535,
title = {MapleGrasp: Mask-guided Feature Pooling for Language-driven Efficient Robotic Grasping},
author = {Vineet Bhat and Naman Patel and Prashanth Krishnamurthy and Ramesh Karri and Farshad Khorrami},
journal= {arXiv preprint arXiv:2506.06535},
year = {2025}
}