We introduce RF-Behavior, a multimodal radio frequency dataset for comprehensive human activity and affective behavior analysis. We collected data from 61 participants performing 21 gestures, 10 activities, and 6 affective behavior expressions from three environmental settings: lab, living room and industrial factory. Data were captured using synchronized sensors, including 13 radars (8 ground-mounted and 5 ceiling-mounted), 6 to 8 RFID tags (attached to arms and torso) and a LoRa transmitter/receiver. Inertial measurement units (IMUs) and 24 infrared cameras are used to provide precise motion ground truth. RF-Behavior comprises a unified multimodal dataset spanning a wide spectrum of human behavior, from brief gestures to activities and affective behaviors, enabling research on multi-task learning across motion and affective behavior recognition. Benchmark results demonstrate that the strategic sensor placement is complementary across modalities, with distinct performance characteristics for different behavioral categories.

翻译：暂无翻译