用关节角度实现工业场景下实时人体动作识别，更稳定且适应性强。
Biomechanically consistent real-time action recognition for human-robot interaction
- 基于关节角度而非关节点位置，引入生物力学先验提升鲁棒性。
- 在11名受试者数据上达88%准确率，对非正对摄像头的姿势也有效。
- 适合机器人实时交互场景，可直接部署于标准2D摄像头系统。
本文提出一种基于标准2D摄像头的工业场景下实时人体动作识别新框架。构建了完整的关节运动学估计流水线，输入至时序平滑的Transformer网络进行动作识别。使用包含11名受试者执行多种动作的新数据集评估方法。与多数依赖关节点位置（JCP）且为离线处理的研究不同，本方法采用关节角度作为输入，利用生物力学先验，实现快速、鲁棒的实时识别。关节角度使方法对传感器姿态、受试者姿态及体型差异具有不变性，并确保跨环境与受试者的泛化能力。所提学习模型在多个指标上优于最佳基线模型，运行于实时系统，达到88%准确率，且对未正对摄像头的受试者仍具良好表现。通过在线人机交互实验验证了技术的鲁棒性与实用性，模拟机器人可实时根据识别动作做出响应。
原文摘要 · Abstract (English)
This paper presents a novel framework for real-time human action recognition in industrial contexts, using standard 2D cameras. We introduce a complete pipeline for robust and real-time estimation of human joint kinematics, input to a temporally smoothed Transformer-based network, for action recognition. We rely on a new dataset including 11 subjects performing various actions, to evaluate our approach. Unlike most of the literature that relies on joint center positions (JCP) and is offline, ours uses biomechanical prior, eg. joint angles, for fast and robust real-time recognition. Besides, joint angles make the proposed method agnostic to sensor and subject poses as well as to anthropometric differences, and ensure robustness across environments and subjects. Our proposed learning model outperforms the best baseline model, running also in real-time, along various metrics. It achieves 88% accuracy and shows great generalization ability, for subjects not facing the cameras. Finally, we demonstrate the robustness and usefulness of our technique, through an online interaction experiment, with a simulated robot controlled in real-time via the recognized actions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。