ScalAR: Authoring Semantically Adaptive Augmented Reality Experiences in Virtual Reality

Fast-Forward Reality: Authoring Error-Free Context-Aware Policies with Real-Time Unit Tests in Extended Reality

“…Moreover, we would discuss the possibility of deploying one CAP across different environments, which has been addressed by prior authoring systems [42,72,100]. Following the current context factor framework, one straightforward answer is that the CAP could be correctly run as long as all the involving context instances are present in the new environment.…”

“…Similarly, this concern has been addressed in the research area of adaptive systems. ScalAR [72] adopts the Genetic Algorithm to enable adaptive AR application authoring within a small scale of indoor layouts. It considers how to generate the most representative and diversified scenes with spatial and identity variations to help AR designers efficiently validate the AR element behaviors and identify failure cases.…”

“…Further, while being immersed in XR environments, end-users can experience simulated scenarios free from the limitations of physical contexts such as user status and time. Users can revisit what happened in the past via in-situ XR visualizations of human activities [13,30,52] and interactions [73,99], or virtually immerse into other times and spaces [72,94,108]. Finally, through immersive authoring [45], users can intuitively program and author XR applications from within the target deployment environment [38,65,66,105], which enables a real-time and seamless transition between authoring and testing.…”

XAIR: A Framework of Explainable AI in Augmented Reality

“…With 3D sensing capabilities, the location of an explanation can be body-based (mostly explicit), object-based (implicit or explicit), or world-based (implicit or explicit) [35,120,145,227]. Prior AR research has explored adaptive interface locations [147,157], e.g., interfaces should be adaptive based on the semantic understanding of the ongoing interaction [54,171,185] and ergonomic metrics [79].…”

A meta-framework for multi-label active learning based on deep reinforcement learning

“…Lindlbauer et al [31] control the placement and level of detail of virtual content by considering both the indoor environment and the cognitive load of the user's performing task. ScalAR [40] enables designers to author semantically adaptive AR experiences in VR. However, all these works focus on static scenarios where the physical objects are placed in fixed positions.…”