MMEAD-VRAG: A Multi-Modal Continuous Emotion Annotation Dataset for VR Action Games

Furthermore, we introduce A Multi-Modal Continuous Emotion Annotation Dataset for VR Action Games (MMEAD-VRAG), the first multi-modal time-series dataset incorporating both physiological and behavioral signals in VR action gaming scenarios. A comparative analysis with existing state-of-the-art datasets reveals that MMEAD-VRAG exhibits fewer limitations in terms of data collection methodology, dataset scale, and participant diversity. The implementation of PhyBehavNet and the MMEAD-VRAG dataset is publicly available at https://github.com/EnyaoC/MMEAD-VRAG.

此外，本文提出面向VR动作游戏的多模态连续情感标注数据集（MMEAD-VRAG），这是首个在VR动作游戏场景中融合生理信号与行为信号的多模态时序数据集。通过与现有前沿数据集开展对比分析可知，MMEAD-VRAG在数据采集方法、数据集规模以及参与者多样性维度上的局限性均更少。PhyBehavNet的实现代码及MMEAD-VRAG数据集已在https://github.com/EnyaoC/MMEAD-VRAG公开。