I-Corps: Privacy-Responsive Artificial Intelligence-Based Solution for Smart Video Surveillance

这个i-Corps项目的更广泛的影响/商业潜力正在探索社会需求、市场前景以及利用最新的人工智能突破来应对公共安全挑战方面的障碍。该项目揭示并采用最佳实践，以最大限度地发挥智能视频监控技术的优势，同时解决相关的隐私和伦理问题。该项目旨在创建一种负责任和隐私响应的人工智能，各种利益相关者、企业主、当地社区和公众都可以接受、采用和使用。通过以道德和负责任的方式使用先进技术，该项目旨在在当局和平民之间建立缓冲，最大限度地减少不必要的冲突，并消除基于种族、民族、性别、年龄和社会经济地位的偏见。该i-Corps项目基于开发基于人工智能的计算机视觉和视频分析技术，以执行复杂的认知任务，以识别社区空间中潜在的威胁异常行为。此外，它将通过提供移动应用程序和Web门户来通知居民、通勤者和负责的安全人员，从而使最终用户能够查看基于时间事件和统计数据的荟萃分析。整个智能处理将在摄像机旁执行，而无需将实际视频保存或发送到外部服务器。对于异常检测，拟议的技术将不使用个人身份信息或面部识别，仅依赖社区空间内的行为和活动。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this I-Corps project is exploring societal needs, market prospects, and obstacles in harnessing the latest artificial intelligence breakthroughs to tackle public safety challenges. This project uncovers and adopts optimal practices to maximize the advantages of intelligent video surveillance technologies while addressing the associated privacy and ethical considerations. This project seeks to create a responsible and privacy-responsive artificial intelligence that various stakeholders, business owners, local communities, and the public can accept, adopt, and use. By using advanced technologies ethically and responsibly, this project aims to create a buffer between authorities and civilians, minimizing unnecessary conflicts and eliminating biases based on race, ethnicity, gender, age, and socioeconomic status.This I-Corps project is based on developing artificial intelligence-based computer vision and video analytic technologies to perform complex cognitive tasks to identify potentially threatening anomalous behaviors in community spaces. Furthermore, it will equip end-users to view meta-analyses based on temporal events and statistics by providing a mobile application and a web portal to notify the residents, commuters, and responsible security personnel. The entire intelligent processing will perform next to the cameras without saving or sending the actual videos to an external server. For anomaly detection, the proposed technology will not use personally identifiable information or facial recognition and only relies on behaviors and activities within community spaces.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.