Computer Graphics

这项拟议的研究将开发将室内3D结构与人类行为联系起来的算法。捕捉这种联系对于识别和预测人与物体的交互方式以及实现室内和人类行为的交互式虚拟3D表示的高效设计和生成都是有用的。这些3D表示被用于建筑可视化、游戏和电影的视觉效果制作、虚拟和增强现实、人体工程学分析，以及3D场景理解和人工智能研究。最终，生成以人类为中心的3D场景的算法将在这些领域产生巨大的经济和社会影响。

The proposed research will develop algorithms that connect the 3D structure of interiors with human actions. Capturing this connection is useful for recognizing and predicting how people interact with objects, and for enabling efficient design and generation of interactive virtual 3D representations of interiors and human actions. These 3D representations are used in architectural visualization, visual effects production for games and films, virtual and augmented reality, ergonomics analysis, as well as 3D scene understanding and AI research. Ultimately, algorithms that generate human-centric 3D scenes will have tremendous economic and social impact in these domains.