Situated visualization has become an increasingly popular research area in the visualization community, fueled by advancements in augmented reality (AR) technology and immersive analytics. Visualizing data in spatial proximity to their physical referents affords new design opportunities and considerations not present in traditional visualization, which researchers are now beginning to explore. However, the AR research community has an extensive history of designing graphics that are displayed in highly physical contexts. In this work, we leverage the richness of AR research and apply it to situated visualization. We derive design patterns which summarize common approaches of visualizing data in situ. The design patterns are based on a survey of 293 papers published in the AR and visualization communities, as well as our own expertise. We discuss design dimensions that help to describe both our patterns and previous work in the literature. This discussion is accompanied by several guidelines which explain how to apply the patterns given the constraints imposed by the real world. We conclude by discussing future research directions that will help establish a complete understanding of the design of situated visualization, including the role of interactivity, tasks, and workflows.

翻译：情境可视化已成为可视化社区中日益热门的研究领域，这得益于增强现实技术和沉浸式分析的发展。在数据与其物理参照物空间邻近处进行可视化，带来了传统可视化中不存在的全新设计机遇和考量，研究人员现已开始探索这些方面。然而，增强现实研究领域在设计高度物理环境下显示的图形方面拥有丰富历史。在本工作中，我们利用增强现实研究的丰富资源，将其应用于情境可视化。我们推导出总结数据原位可视化常见方法的设计模式。这些模式基于对AR和可视化社区发表的293篇论文的调查，以及我们自身的专业知识。我们讨论了有助于描述我们的模式和文献中先前工作的设计维度。此讨论附带了若干指南，解释了在现实世界施加的约束条件下如何应用这些模式。最后，我们探讨了未来研究方向，这些方向将有助于建立对情境可视化设计的全面理解，包括交互性、任务和工作流的作用。