Abstract
We present RetailOpt, a novel opt-in, easy-to-deploy system for tracking customer movements in indoor retail environments. The system utilizes information presently accessible to customers through smartphones and retail apps: motion data, store map, and purchase records. The approach eliminates the need for additional hardware installations/maintenance and ensures customers maintain full control of their data. Specifically, RetailOpt first employs inertial navigation to recover relative trajectories from smartphone motion data. The store map and purchase records are then cross-referenced to identify a list of visited shelves, providing anchors to localize the relative trajectories in a store through continuous and discrete optimization. We demonstrate the effectiveness of our system through systematic experiments in five diverse environments. The proposed system, if successful, would produce accurate customer movement data, essential for a broad range of retail applications, including customer behavior analysis and in-store navigation. The potential application could also extend to other domains such as entertainment and assistive technologies.
Abstract (translated)
我们提出了RetailOpt，一种新型的opt-in，易于部署的系统，用于在室内零售环境中跟踪顾客的运动。该系统利用现有顾客通过智能手机和零售应用程序可获得的信息：运动数据、商店地图和购买记录。这种方法消除了额外硬件安装/维护的需求，并确保客户对他们的数据保持完全控制。具体来说，RetailOpt首先采用惯性导航从智能手机运动数据中恢复相对轨迹。商店地图和购买记录随后交叉参考，以确定已访问的货架列表，提供通过连续和离散优化在商店中定位相对轨迹的锚点。我们在五种不同的环境中进行了系统实验，以证明我们系统的有效性。如果成功，该系统将产生准确的客户运动数据，这是广泛的零售应用程序所必需的，包括客户行为分析和在店导航。该系统的潜在应用还可能扩展到其他领域，如娱乐和辅助技术。
URL
https://arxiv.org/abs/2404.12548
