Diverse R-PPG: Camera-Based Heart Rate Estimation for Diverse Subject Skin-Tones and Scenes

Posted:

Pradyumna Chari1, Krish Kabra1, Doruk Karinca1, Soumyarup Lahiri1, Diplav Srivastava1, Kimaya Kulkarni1, Tianyuan Chen1, Maxime Cannesson2, Laleh Jalilian2, Achuta Kadambi1,3,*

1Department of Electrical and Computer Engineering, UCLA, Los Angeles, California 90095, USA
2Department of Anesthesiology and Perioperative Medicine, UCLA, Los Angeles, California 90095, USA
3California NanoSystems Institute, University of California, Los Angeles, California 90095, USA
*Corresponding author: achuta@ee.ucla.edu

Website arXiv PDF

Abstract

Heart rate (HR) is an essential clinical measure for the assessment of cardiorespiratory instability. Since communities of color are disproportionately affected by both COVID-19 and cardiovascular disease, there is a pressing need to deploy contactless HR sensing solutions for high-quality telemedicine evaluations. Existing computer vision methods that estimate HR from facial videos exhibit biased performance against dark skin tones. We present a novel physics-driven algorithm that boosts performance on darker skin tones in our reported data. We assess the performance of our method through the creation of the first telemedicine-focused remote vital signs dataset, the VITAL dataset. 432 videos (~864 minutes) of 54 subjects with diverse skin tones are recorded under realistic scene conditions with corresponding vital sign data. Our method reduces errors due to lighting changes, shadows, and specular highlights and imparts unbiased performance gains across skin tones, setting the stage for making medically inclusive non-contact HR sensing technologies a viable reality for patients of all skin tones.

Citation

@inproceedings{chari2020-diverse-rppg,
    title={Diverse R-PPG: Camera-Based Heart Rate Estimation for Diverse Subject Skin-Tones and Scenes}, 
    author={Pradyumna Chari and Krish Kabra and Doruk Karinca and Soumyarup Lahiri and Diplav Srivastava and Kimaya Kulkarni and Tianyuan Chen and Maxime Cannesson and Laleh Jalilian and Achuta Kadambi},
    year={2020}, 
    eprint={2010.12769},
    archivePrefix={arXiv},
    primaryClass={eess.IV}
}
Direct Link