Mixed reality gesture based interaction for immersive analytics

点击翻译按钮获取中文摘要

NOVA Chemicals has a need to understanding and exploring vast amounts of multi-dimensional (engineering and product quality) data to make better decisions and improve their chemical processes. Understanding the data is an immense problem, requiring advanced statistics & machine learning as well as visualization to bring humans into the decision making loop. The data sets are multi-dimensional in nature and often require a reduction of the dimensionality to two before a visual representation can be created. The project goal is to investigate if immersive approaches (augmented reality, mixed reality, virtual reality) to visualize and interact with these data sets can enhance data exploration and understanding. Immersive analytics visualization tools would be coupled with variable reduction techniques such as Principle Components Analysis, Partial Least Squares and Multi-way Partial Least Squares, to identify interesting high-dimensional patterns in the data as a first step in leveraging this information to improve NOVA products and its manufacturing efficiency. Concretely, we will investigate if gesture interactions can be used effectively to interact with chemical process data and if the field of view of immersive devices has an impact on human task performance.