Self-Organizing Maps for Intuitive Gesture-Based Geometric Modelling in Augmented Reality
2018 IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR)(2018)
摘要
Modelling three-dimensional virtual objects in the context of architectural, product and game design requires elaborate skill in handling the respective CAD software and is often tedious. We explore the potentials of Kohonen networks, also called self-organizing maps (SOM) as a concept for intuitive 3D modelling aided through mixed reality. We effectively provide a computational "clay" that can be pulled, pushed and shaped by picking and placing control objects with an augmented reality headset. Our approach benefits from combining state of the art CAD software with GPU computation and mixed reality hardware as well as the introduction of custom SOM network topologies and arbitrary data dimensionality. The approach is demonstrated in three case studies.
更多查看译文
关键词
augmented reality,computer-aided design,parametric design,self-organizing maps
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要