研究人员开发了一个名为“上下文模型预测生成”（ICMPG）的新框架，以改进从文本描述合成人类运动。该方法结合了大语言模型（LLMs）的语义理解和模型预测控制的物理真实性。ICMPG使用LLM来规划和生成运动序列，然后通过物理模拟和语义对齐进行优化，创建一个适应指令和物理约束的闭环系统。实验表明，与现有方法相比，ICMPG生成的运动在物理上更合理，语义上也更忠实。 AI
影响 通过弥合语言理解与物理模拟之间的差距，该框架有可能实现更逼真、更可控的数字虚拟形象和虚拟环境。
排序理由 该集群描述了一篇详细介绍新颖运动合成框架的研究论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2606.26981v1 Announce Type: cross Abstract: Synthesizing human motion from textual descriptions is essential for immersive digital applications, yet existing methods face a persistent trade-off between semantic fidelity and physical realism. Large language model (LLM)-based…
Synthesizing human motion from textual descriptions is essential for immersive digital applications, yet existing methods face a persistent trade-off between semantic fidelity and physical realism. Large language model (LLM)-based approaches can interpret diverse open-vocabulary …
Synthesizing human motion from textual descriptions is essential for immersive digital applications, yet existing methods face a persistent trade-off between semantic fidelity and physical realism. Large language model (LLM)-based approaches can interpret diverse open-vocabulary …