HRI-AVC

HRI-AVC数据集是由罗切斯特理工学院创建的，用于研究人机交互中的情感估计。该数据集包含在人机协作装配任务中收集的视频和主观测量数据，其中参与者在每次任务迭代后自我报告情感状态（如愉悦度和激动度）。数据集的特点在于情感标签是自我报告的，而非由第三方注释，这使得数据更贴近自然情境。数据集的创建过程涉及复杂的实验设计，包括机器人行为的多种参数调整以诱发不同的情感反应。HRI-AVC数据集主要应用于情感计算和计算机视觉领域，旨在解决在人机交互中实时估计用户情感状态的问题。

The HRI-AVC dataset was created by Rochester Institute of Technology for research on emotion estimation in human-robot interaction. This dataset comprises video and subjective measurement data collected during human-robot collaborative assembly tasks, where participants self-reported their emotional states (e.g., valence and arousal) after each task iteration. A core characteristic of this dataset is that its emotional labels are self-reported rather than third-party annotated, which renders the data more aligned with naturalistic real-world scenarios. The dataset's development involved sophisticated experimental design, including the tuning of multiple parameters of robot behaviors to elicit diverse emotional responses from participants. The HRI-AVC dataset is primarily utilized in the fields of affective computing and computer vision, with the goal of addressing the challenge of real-time estimation of user emotional states during human-robot interaction.