Quantifying Gaze Behavior During Real-World Interactions Using Automated Object, Face, and Fixation Detection
Quantifying Gaze Behavior During Real-World Interactions Using Automated Object, Face, and Fixation Detection
复制标题
DOI：
10.1109/tcds.2018.2821566
复制
发表时间：
2018-03
影响因子：
5
通讯作者：
L. Chukoskie;Shengyao Guo;Eric Ho;Yalun Zheng;Qiming Chen;Vivian Meng;John Cao;Nikhita Devgan;Si Wu;P. Cosman
中科院分区：
文献类型：
--
作者：
L. Chukoskie;Shengyao Guo;Eric Ho;Yalun Zheng;Qiming Chen;Vivian Meng;John Cao;Nikhita Devgan;Si Wu;P. Cosman
随着用于在真实的世界社交环境中获取注视行为的技术的发展，需要使受过训练的观察者编码行为所需的时间最小化的鲁棒方法。我们记录了一个戴着眼动跟踪眼镜的受试者在与其他三个人进行自然交互时的凝视行为，在交互过程中涉及或操纵多个对象。每个交互产生的凝视世界视频可以针对不同的行为进行手动编码，但这非常耗时，并且需要经过培训的行为编码人员。相反，我们使用神经网络来检测对象，并使用具有特征跟踪的Viola-Jones框架来检测人脸。的时间序列的凝视降落在对象/面部边界框内的行程长度，以确定“外观”，我们讨论的运行长度参数的优化。算法性能进行比较，对专家整体地面真相。
As technologies develop for acquiring gaze behavior in real world social settings, robust methods are needed that minimize the time required for a trained observer to code behaviors. We record gaze behavior from a subject wearing eye-tracking glasses during a naturalistic interaction with three other people, with multiple objects that are referred to or manipulated during the interaction. The resulting gaze-in-world video from each interaction can be manually coded for different behaviors, but this is extremely time-consuming and requires trained behavioral coders. Instead, we use a neural network to detect objects, and a Viola–Jones framework with feature tracking to detect faces. The time sequence of gazes landing within the object/face bounding boxes is processed for run lengths to determine “looks,” and we discuss optimization of run length parameters. Algorithm performance is compared against an expert holistic ground truth.