Abstract
Wearable Internet of Things (IoT) devices, such as VR/XR head-mounted displays, enable immersive interaction with complex data that is commonly organized in a tabular format. However, precisely and efficiently selecting tabular data in immersive environments is challenging due to input instability and spatial tracking limitations. We introduce iSelecTable, a framework designed for the interactive selection of tabular data in virtual reality. The framework features three selection methods — click, open-path, and close-path — for three interaction modes: controller, bare-hand, and gaze-pinch interactions. The framework consists of different gestures to select single or multiple data regions, including rows, columns, cells, and continuous blocks, with refinement mechanisms to correct boundary errors. Under the evaluated device and controlled experimental configuration, our evaluation demonstrates that selection methods show distinct suitability across granularities: click for precise cells, open-path for larger rows or blocks, and close-path degrades more gradually under controlled path perturbations, supporting robust free exploration in immersive settings. Regarding the performance and physical workload, controllers maximize efficiency, while gaze-pinch offers a lighter but slightly less precise alternative. These insights highlight that balancing accuracy with user effort is crucial for future tabular interaction design, guiding adaptive mechanisms for immersive tabular data exploration.