arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.03672cs.CV
收藏

基于场景上下文的长时间人体运动预测

Long-term Human Motion Prediction with Scene Context

UC Berkeley(加州大学伯克利分校);Nanjing University(南京大学)

Zhe Cao, Hang Gao, Karttikeya Mangalam, Qi-Zhi Cai, Minh Vo, Jitendra Malik

摘要:

人的运动是目标导向的,并受到场景中对象的空间布局的影响。为了规划未来的人类运动,感知环境是至关重要的-想象一下,在一个关灯的新房间里导航是多么困难。现有的预测人体运动的工作不注意的场景上下文,从而在长期预测的斗争。在这项工作中,我们提出了一个新的三阶段框架,利用场景上下文来解决这个任务。给定一幅场景图像和2D姿态历史,我们的方法首先采样多个人体运动目标,然后规划朝向每个目标的3D人体路径,最后预测沿着每个路径的3D人体姿态序列。为了稳定的训练和严格的评估,我们提供了一个具有干净注释的多样化合成数据集。在合成和真实数据集,我们的方法显示出一致的定量和定性改进现有的方法。

英文摘要:

Human movement is goal-directed and influenced by the spatial layout of the objects in the scene. To plan future human motion, it is crucial to perceive the environment -- imagine how hard it is to navigate a new room with lights off. Existing works on predicting human motion do not pay attention to the scene context and thus struggle in long-term prediction. In this work, we propose a novel three-stage framework that exploits scene context to tackle this task. Given a single scene image and 2D pose histories, our method first samples multiple human motion goals, then plans 3D human paths towards each goal, and finally predicts 3D human pose sequences following each path. For stable training and rigorous evaluation, we contribute a diverse synthetic dataset with clean annotations. In both synthetic and real datasets, our method shows consistent quantitative and qualitative improvements over existing methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑