CRI: II-New: Cognitive Mechanisms and Computational Modeling of Gaze Control During Scene Free Viewing, Visual Search, and Daily Tasks

CRI：II-新：场景自由观看、视觉搜索和日常任务期间注视控制的认知机制和计算模型

• 批准号: 1823276
• 负责人: Mubarak Shah
• 金额: $ 24.93万
• 依托单位: The University of Central Florida Board of Trustees
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-08-15 至 2021-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1823276&HistoricalAwards=false
• 关键词: CRI; II; New; Cognitive; Mechanisms

This project is to build an infrastructure to study visual attention and its applications at the University of Central Florida. The infrastructure will also enable larger-scale inquiries into high-level visual recognition problems in the vision sciences. The infrastructure of fixed and portable eye trackers will be used by several faculty and researchers across the University (e.g., Computer Science, Engineering and Computer Engineering, Psychology, Biomedical engineering and UCF Institute for Simulation and Training) to study human perception, cognition, learning, and motor control, as well as exploring applications in activity recognition, surveillance, and data summarization. The core goal of this infrastructure is to utilize gaze and eye tracking technology to understand mechanisms of attention from behavioral, neurophysiological, and computational perspectives and explore its applications in a wide range of problems in computer vision and psychology. The team utilizing the infrastructure has complementary expertise to address research encompassing computer vision, machine learning, human vision, imaging, and psychology. In particular, this infrastructure will be used to explore in what ways current attention models fail, how to remedy them, and discover new cues that attract gaze. The approach is a combination of cognitive and computational studies utilizing machine learning and computer vision techniques. New large-scale eye movement datasets and benchmarks will be constructed as part of this proposal.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该项目旨在在中佛罗里达大学建立一个研究视觉注意力及其应用的基础设施。该基础设施还将支持对视觉科学中的高级视觉识别问题进行更大规模的调查。固定式和便携式眼动仪的基础设施将被大学的几位教师和研究人员（例如计算机科学、工程学和计算机工程、心理学、生物医学工程和 UCF 模拟与培训研究所）使用，以研究人类感知、认知、学习和运动控制，并探索活动识别、监视和数据汇总方面的应用。该基础设施的核心目标是利用凝视和眼球追踪技术从行为、神经生理学和计算角度理解注意力机制，并探索其在计算机视觉和心理学领域广泛问题中的应用。利用该基础设施的团队拥有互补的专业知识，可以解决计算机视觉、机器学习、人类视觉、成像和心理学等领域的研究。特别是，该基础设施将用于探索当前注意力模型在哪些方面失败，如何补救，并发现吸引目光的新线索。该方法是利用机器学习和计算机视觉技术的认知和计算研究的结合。作为该提案的一部分，将构建新的大规模眼动数据集和基准。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。