CAREER: Enabling Seamless Vision Sensing in Cloud-Edge Systems

职业：在云边缘系统中实现无缝视觉传感

• 批准号: 1846320
• 负责人: Karthik Dantu
• 金额: $ 54.94万
• 依托单位: SUNY at Buffalo
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-06-01 至 2024-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1846320&HistoricalAwards=false
• 关键词: CAREER; Enabling; Seamless; Vision; Sensing

Mobile devices such as smartphones have changed our daily lives. With the growth in the capabilities of smartphones as well as wearables such as "smart glasses", the next decade promises richer capabilities in terms of augmented reality, improved contextual sensing and more seamless computing that will integrate into our day-to-day lives. However, vision sensing applications, which are fueled by the improved understanding of user context, are highly resource and energy intensive, reducing useful operation time.This project will enable the design of such applications by providing a programming framework that allows future applications to leverage cloud computation seamlessly, to provide the user with a good quality of service for the intended application while removing from programmers the burden of reasoning about where the computation should happen. The project includes plans for integration of research in education, broadening participation, and educational outreach.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.

智能手机等移动设备改变了我们的日常生活。随着智能手机以及“智能眼镜”等可穿戴设备功能的增长，未来十年有望在增强现实、改进的情境感知和更无缝的计算方面提供更丰富的功能，这些功能将融入我们的日常生活。然而，通过提高对用户上下文的理解而推动的视觉传感应用程序是高度资源和能源密集型的，从而减少了有用的操作时间。该项目将通过提供一个编程框架来实现此类应用程序的设计，该框架允许未来的应用程序无缝地利用云计算，为用户的预期应用程序提供良好的服务质量，同时消除程序员推理应该在哪里进行计算的负担。 该项目包括将研究融入教育、扩大参与和教育推广的计划。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。

项目成果

A quantitative analysis of system bottlenecks in visual SLAM

• DOI: 10.1145/3508396.3512882
• 发表时间: 2022-03
• 期刊: Proceedings of the 23rd Annual International Workshop on Mobile Computing Systems and Applications
• 作者: S. Semenova;Steven Y. Ko;Yu David Liu;Lukasz Ziarek;Karthik Dantu

Practical Persistence Reasoning in Visual SLAM

• DOI: 10.1109/icra40945.2020.9196913
• 发表时间: 2020-05
• 期刊: 2020 IEEE International Conference on Robotics and Automation (ICRA)
• 作者: Z. S. Hashemifar;Karthik Dantu

Edge-SLAM: Edge-Assisted Visual Simultaneous Localization and Mapping

• DOI: 10.1145/3561972
• 发表时间: 2020-06
• 期刊: ACM Transactions on Embedded Computing Systems
• 影响因子: 2
• 作者: Ali J. Ben Ali-Ali-J.-Ben-Ali-46653644;Z. S. Hashemifar;Karthik Dantu