CardiacMamba uses state space modeling for robust heart rate estimation

Researchers have developed CardiacMamba, a novel framework for estimating heart rate from facial videos using a fusion of RGB and radio-frequency (RF) data. This approach leverages state space modeling to enhance the accuracy and robustness of heart rate monitoring, particularly under challenging conditions like varying illumination and RF signal loss. CardiacMamba demonstrates state-of-the-art performance on the EquiPleth dataset, significantly reducing errors and bias across different skin tones. AI

IMPACT This research advances non-contact physiological monitoring by improving the accuracy and fairness of heart rate estimation from video data.

RANK_REASON The cluster contains a research paper detailing a new method for heart rate estimation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• CardiacMamba
• Channel-wise Fast Fourier Transform
• EquiPleth
• RGB-RF
• Temporal Difference Mamba Module

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →