Gaze scanning by walking people with visual field loss

项目摘要 在移动任务中，我们通常通过移动眼睛和头部来环顾周围的环境。由于眼球运动 幅度通常较小（<15º），头部运动对凝视扫描的贡献可能较大， 更重要对于行走时视野丧失的患者，我们以前的研究没有发现他们的眼睛 运动比正常情况下分散得更广。剩下的一个问题是， 通过移动他们的头。然而，由于头部跟踪的技术挑战， 户外行走场景中的运动行为比眼睛运动相对稀疏。过几 包括头部跟踪的研究，调查要么在虚拟环境中进行，要么 仅定性评估头部运动。 在这个项目中，我们建议对以前联邦资助的项目中收集的数据进行二次分析。 研究项目，在这些项目中，我们用头戴式摄像机记录了头中的眼睛和场景视频， 当他们走在波士顿忙碌的街道上时， 市中心。为了从场景视频中提取头部运动，我们将使用最新的3D视觉映射 技术，近年来随着应用于自动驾驶研究而取得了实质性进展 如今。这将是第一次完整的凝视扫描模式，包括头部运动， 对户外开放环境中的视力障碍者进行定量评估。 先前收集的视频还包括一个干预条件，其中半野丢失的患者 戴着棱镜场扩展器行走。通过比较有和没有 如果我们能够扩展视野，我们就能够理解如何干预凝视扫描行为。的 从这项研究中获得的知识可能会为低视力康复社区提供信息。

Project Summary In mobility tasks, we look around the environment usually by moving both eye and head. Since eye movement amplitude is usually small (<15º), the contribution of head movement to the gaze scanning may be larger and more important. For patients with visual field loss in walking, our previous studies did not find their eye movements dispersed wider than normally sighted. A remaining question is whether the patients primarily scan by moving their heads. However, due to the technically challenges in head tracking, research on head movement behaviors in outdoor walking scenarios is relatively sparse than that for eye movement. In a few studies that do include head tracking, the investigation was either conducted in virtual environment, or the head movements were only qualitatively assessed. In this project, we propose to conduct secondary analysis of data collected in previous federally funded research projects, in which we recorded the eye-in-head and scene videos with a head mounted camera in patients with visual field loss as well as normally sighted controls, as they walked on busy streets in Boston downtown area. To extract head movements from the scene videos, we will use the latest 3D visual mapping technology, which has substantially advanced in recent years as it is been applied to self-driving research nowadays. This would be the first time the complete gaze scanning patterns including the head movements of the visually impaired people in an open outdoor environment are quantitatively evaluated. The previously collected videos also include an intervention condition, where the patients with hemi-field loss walking while wearing a prism field expander. By comparing the gaze scanning behavior with and without the field expander, we would be able to understand how gaze scanning behaviors can be intervened. The knowledge gained from this research can potentially inform the low vision rehabilitation community.