Opt-in Camera: Person Identification in Video via
UWB Localization and Its Application to Opt-in Systems

This paper presents opt-in camera, a concept of privacy-preserving camera systems capable of recording only specific individuals in a crowd who explicitly consent to be recorded. Our system utilizes a mobile wireless communication tag attached to personal belongings as proof of opt-in and as a means of localizing tag carriers in video footage. Specifically, the on-ground positions of the wireless tag are first tracked over time using the unscented Kalman filter (UKF). The tag trajectory is then matched against visual tracking results for pedestrians found in videos to identify the tag carrier. Technically, we devise a dedicated trajectory matching technique based on constrained linear optimization, as well as a novel calibration technique that handles wireless tag-camera calibration and hyperparameter tuning for the UKF, which mitigates the non-line-of-sight (NLoS) issue in wireless localization. We realize the proposed opt-in camera system using ultra-wideband (UWB) devices and an off-the-shelf webcam installed in the environment. Experimental results demonstrate that our system can perform opt-in recording of individuals in near real-time at 10~fps, with reliable identification accuracy for a crowd of 8-23 people in a confined space.

Cameras are prevalent in a variety of robotics applications, such as mobile robots, social robots, and intelligent surveillance systems. Yet, when it comes to the practical deployment of such camera-equipped robots, privacy and personal data protection pose significant challenges. The EU's General Data Protection Regulation (GDPR) mandates that business stakeholders obtain valid consent from individuals when processing their personal data, including visual features that can lead to personal identification, specifically for direct marketing applications. Even when not legally required, obtaining informed consent can often be best practice from an ethical perspective, especially in sensitive settings like hospitals and homes. Given these considerations, we introduce the concept of privacy-preserving camera systems named opt-in camera, which records individuals in a crowd only when they have explicitly provided their consent to be recorded (i.e., opt-in).

The figure above provides an overview of our opt-in camera system. We assume that individuals declare their opt-in for data recording by carrying small UWB tag devices. Our system identifies those tag-carrying, opt-in individuals in camera footage and mask the other people by replacing their regions with background images. This procedure consists of the following four steps:

1. tracking: to estimate on-ground trajectories of the tag devices
2. camera-based tracking: to obtain a collection of tracklets for pedestrians shown in a video
3. trajectory matching: to identify tag carriers in the video
4. masking: the video so as to contain the only opt-in individuals

In addition, a dedicated system calibration is proposed to improve tag-carrier identification accuracy.