English(EN) Evaluating the Performance of Deep Learning Models in Whole-body Dynamic 3D Posture Prediction During Load-reaching Activities

深度学习模型预测人体在动态负重抓取任务中的姿态

作者 PulseAugur 编辑部 · [1 个来源] · 2026-06-02 04:00

研究人员开发了深度学习模型，特别是BLSTM和Transformer架构，用于预测人体在动态负重抓取活动中的身体姿态。模型利用手部负重位置、抓取技术和初始身体姿态数据来预测后续动作。引入了一种新颖的成本函数来强制身体各部分长度恒定，将预测精度提高了21%。Transformer模型表现出优越的性能，均方根误差达到41.4毫米，在长期预测方面比BLSTM模型优越约58%。 AI

影响为手动物料搬运中人体运动动力学预测引入了改进的AI方法。

排序理由这是一篇研究论文，详细介绍了AI模型在特定任务中的新颖方法和结果。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Seyede Niloofar Hosseini, Ali Mojibi, Mahdi Mohseni, Navid Arjmand, Alireza Taheri · 2026-06-02 04:00

Evaluating the Performance of Deep Learning Models in Whole-body Dynamic 3D Posture Prediction During Load-reaching Activities

arXiv:2511.20615v2 Announce Type: replace-cross Abstract: This study aimed to explore the application of deep neural networks for whole-body human posture prediction during dynamic load-reaching activities. Two time-series models were trained using bidirectional long short-term m…

报道来源 [1]

Evaluating the Performance of Deep Learning Models in Whole-body Dynamic 3D Posture Prediction During Load-reaching Activities

相关实体

相关话题