研究人员开发了一种新方法，可以直接从无标签的真实视频中训练神经对象动力学模型，克服了合成数据的局限性。该框架使用基于粒子的动力学模型并与高斯飞溅相结合，以预测粒子位置和旋转随时间的变化。这种方法无需显式的粒子级状态标签即可从真实视频中学习，并包含一个约 500 个视频的新数据集，展示了各种对象交互。 AI
影响 通过直接在真实数据上进行训练，能够实现更逼真的物理模拟，从而可能缩小人工智能中的模拟到真实世界的差距。
排序理由 该集群包含一篇详细介绍人工智能模型训练新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.23845v1 Announce Type: new Abstract: Data-driven learning approaches for physics simulation, sometimes referred to as world models, have emerged as promising alternatives to traditional physics simulators due to their differentiable nature. Prior work has demonstrated …
Data-driven learning approaches for physics simulation, sometimes referred to as world models, have emerged as promising alternatives to traditional physics simulators due to their differentiable nature. Prior work has demonstrated impressive results in predicting the motions of …