眼睛凝视和停留Eye-gaze and dwell

“眼睛凝视和停留”交互模型 是眼睛凝视和提交交互模型的特例:The "eye-gaze and dwell" interaction model is a special case of the eye-gaze and commit interaction model:

  1. 注视目标和Look at a target and
  2. 要确认选择目标,请使用辅助显式输入:只是一直看着要选择的目标。To confirm your intention to select the target, use a secondary explicit input: Simply keep looking at the target you would like to select.

“眼睛凝视和停留”交互模型的优势Advantages of the "eye-gaze and dwell" interaction model

如果你正在用手执行某个任务,或者手中拿了工具,则可能无法用手与全息影像交互。When your hands are already occupied with a task or holding tools, using them for interacting with holograms may not be an option. 选择全息影像时,一种解放双手的替代性交互方式是“眼睛凝视和停留”,换言之,“盯着看”。 A hands-free interaction alternative for selecting holograms is "eye-gaze and dwell" or in other words: "look and stare". 通过这种方式,即使是行动严重受限的无法完全转动其头部或身体的用户(例如,在空间严重受限的工作环境中的用户),也可以与全息影像交互。With this approach, even severely constrained users who can't fully turn their heads or bodies can interact with holograms (for example, in a highly confined work environment). 用户只需一直看着要选择的目标,系统就会显示不同的停留反馈来指示过程。The user simply keeps looking at the target they would like to select and different dwell feedback is displayed to indicate the process.

“眼睛凝视和停留”交互模型的难点Challenges of the "eye-gaze and dwell" interaction model

通常情况下,建议只在语音输入和手动输入都不可用的情况下使用基于停留的激活。In general, we recommend to only use dwell-based activations as a last fall-back if neither voice input nor hand input is available. 原因在于不易选择合适的停留时间。The reason is that the choice of dwell time can be tricky. 新手用户可以使用较长的停留时间,专家用户则希望快速且高效地浏览其整个体验。Novice users are ok with longer dwell times, while expert users want to quickly and efficiently navigate through their experiences. 因此,难点在于如何根据用户的具体需求调整停留时间。This leads to the challenge of how to adjust the dwell time to the specific needs of a user. 如果停留时间太短,则会因为全息影像一直在对用户的眼睛凝视进行响应而导致用户感到不适。If the dwell time is too short: The user may feel overwhelmed by having holograms react to their eye-gaze all the time. 如果停留时间太长,则会给用户造成体验太慢且易中断的感觉,因为用户必须长时间盯着目标。If the dwell time is too long: The experience may feel too slow and interruptive as the user has to keep looking at targets for a long time.

设计建议Design recommendations

建议对停留反馈使用双状态方法:We recommend using a two-state approach for dwell feedback:

  1. 初动延迟 :当用户开始看着目标时,系统不应立即响应,否则可能会给用户带来不愉快的体验,导致用户感到不适。Onset delay: When the user starts looking at a target, nothing should immediately happen as this may result in an unpleasant and overwhelming user experience. 此时应启动一个计时器,用于检测用户是在有意盯着该目标,还是仅仅扫了它一眼。Instead start a timer to detect whether the user is intentionally staring at the target or merely glancing over it. 在设定好临近度(即用户在注视而不是瞟着某个大型目标)的情况下,建议将初动时间设置为 150-250 毫秒。We recommend an onset time of 150-250 ms in a given proximity (meaning the user is fixating vs. looking around on a large target).
  2. 启动停留反馈: 在确定用户是在有意盯着该目标以后,请开始显示停留反馈,通知用户停留激活正在启动。Start dwell feedback: After ensuring that the user is intentionally looking at the target, start showing dwell feedback to inform the user that the dwell activation is being initiated.
  3. 持续反馈: 当用户持续看着目标时,显示一个连续进度指示器,这样用户就知道自己必须始终看着目标。Continuous Feedback: While the user keeps looking at the target, show a continuous progress indicator so that the user knows that they have to keep looking at the target. 具体说来,对于眼睛凝视输入,建议一开始显示一个较大的圆圈或球体,然后将其缩小,通过这种方式吸引用户的视觉注意力。 In particular for eye-gaze input, we recommend pulling in the user's visual attention by starting out with a bigger circle or sphere that contracts into a smaller version. 显示一个表示最终状态的指示器(小圆圈)是为了告知用户停留将何时结束。Showing an indicator for the final state (small circle) helps to communicate to the user when the dwell will be finished. 示例图如下所示。An example illustration is shown below.
  4. 完成: 如果用户一直注视该目标(又注视了 650-850 毫秒),则完成停留激活,选择被用户注视的目标。Finish: If the user kept fixating the target (for another 650-850 ms), complete the dwell activation and select the looked-at target.

停留状态

另请参阅See also