Core Concepts
HOOD proposes a robust and real-time method for human presence and out-of-distribution detection using FMCW radar.
Abstract
The study introduces HOOD, a solution for human presence and OOD detection using 60 GHz FMCW radar.
The method relies on a reconstruction-based architecture and works with radar macro and micro range-Doppler images (RDIs).
HOOD aims to accurately detect human presence in the presence or absence of moving and stationary disturbers.
The solution also serves as an OOD detector, identifying moving or stationary clutters as OOD in the absence of humans.
HOOD outperforms state-of-the-art OOD detection methods and is versatile across different hardware environments.
The study includes detailed experiments and evaluations showcasing the effectiveness of HOOD.
Stats
On our dataset collected with a 60 GHz short-range FMCW radar, we achieve an average AUROC of 94.36%.
Quotes
"HOOD aims to accurately detect the presence of humans in the presence or absence of moving and stationary disturbers."