Practical Human Sensing in the Light

TLDR

StarLight reconstructs fine‑grained user skeleton postures in real time by reusing light emitted from ceiling LED panels. It uses a small array of optimally placed photodiodes to passively capture light‑blockage cues from LED panels, aggregates this data to fit 3D skeleton postures, and is deployed with 20 LED panels and 20 photodiodes in a typical office. StarLight reduces intrusive sensors, overcomes furniture blockage, supports mobility, and achieves a 13.6° mean angular error for five joints at 40 FPS, establishing a new unobtrusive sensing paradigm.

Abstract

We present StarLight, an infrastructure-based sensing system that reuses light emitted from ceiling LED panels to reconstruct fine-grained user skeleton postures continuously in real time. It relies on only a few (e.g., 20) photodiodes placed at optimized locations to passively capture low-level visual clues (light blockage information), with neither cameras capturing sensitive images, nor on-body devices, nor electromagnetic interference. It then aggregates the blockage information of a large number of light rays from LED panels and identifies best-fit 3D skeleton postures. StarLight greatly advances the prior light-based sensing design by dramatically reducing the number of intrusive sensors, overcoming furniture blockage, and supporting user mobility. We build and deploy StarLight in a 3.6 m x 4.8 m office room, with customized 20 LED panels and 20 photodiodes. Experiments show that StarLight achieves 13.6 degree mean angular error for five body joints and reconstructs a mobile skeleton at a high frame rate (40 FPS). StarLight enables a new unobtrusive sensing paradigm to augment today's mobile sensing for continuous and accurate behavioral monitoring.