C-MambaPose: A Physics-Informed Complex Mamba Framework for Cross-Environment WiFi Human Pose Estimation
Researchers have developed C-MambaPose, a novel framework for human pose estimation using WiFi signals. This system leverages complex Mamba and Graph Convolutional Network components to interpret WiFi channel state information, focusing on phase dynamics for improved accuracy. C-MambaPose demonstrates superior performance in cross-environment estimations and significantly reduces parameter count compared to existing methods while maintaining comparable model size. AI
IMPACT This framework could advance device-free human sensing capabilities by improving the accuracy and generalizability of WiFi-based pose estimation.