RetailOpt: Opt-In, Easy-to-Deploy Trajectory Estimation from Smartphone Motion Data and Retail Facility Information

TL;DR

The system uses readily accessible information from customer smartphones and retail apps, including motion data, store maps, and purchase records, to recover relative trajectories from smartphone motion data, and first uses inertial navigation to recover relative trajectories from smartphone motion data.

Abstract

We present RetailOpt, a novel opt-in, easy-to-deploy system for tracking customer movements offline in indoor retail environments. The system uses readily accessible information from customer smartphones and retail apps, including motion data, store maps, and purchase records. This eliminates the need for additional hardware installations/maintenance and ensures customers full data control. Specifically, RetailOpt first uses inertial navigation to recover relative trajectories from smartphone motion data. The store map and purchase records are 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 in five diverse environments. The 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.

Figures (5)

• Figure 1: The RetailOpt system leverages smartphone motion data, store map, and purchase records to enable opt-in, easy-to-deploy indoor trajectory estimation.
• Figure 2: Trajectory estimation pipeline. A relative motion trajectory from inertial navigation (blue lines in the figure, while ground-truth trajectories obtained by SLAM are shown in black lines) is first transformed to better align with anchors (orange circles) created from the store map and purchase records. The transformed trajectory is then projected onto the valid space using a Viterbi algorithm, ensuring the final trajectory does not collide with obstacles in the environment.
• Figure 3: Data collection environments.
• Figure 4: Results on our new dataset. The estimated and ground-truth trajectories are visualized with blue and black lines, respectively. Obstacle regions are colored in gray, and anchors are annotated with orange circles.
• Figure 5: Results on Inertial Localization Dataset. The estimated and ground-truth trajectories are visualized with blue and black lines, respectively. Obstacle regions are colored in gray, and anchors are annotated with orange circles.