A Meta-Analysis of the Impact of Skin Type and Gender on Non-contact Photoplethysmography Measurements
Abstract
It is well established that many datasets used for computer vision tasks are not representative and may be biased. The result of this is that evaluation metrics may not reflect real-world performance and might expose some groups (often minorities) to greater risks than others. Imaging photoplethysmography is a set of techniques that enables noncontact measurement of vital signs using imaging devices. While these methods hold great promise for low-cost and scalable physiological monitoring, it is important that performance is characterized accurately over diverse populations. We perform a meta-analysis across three datasets, including 73 people and over 400 videos featuring a broad range of skin types to study how skin types and gender affect the measurements. While heart rate measurement can be performed on all skin types under certain conditions, we find that average performance drops significantly for the darkest skin type. We also observe a slight drop in the performance for females. We compare supervised and unsupervised learning algorithms and find that skin type does not impact all methods equally. The imaging photoplethysmography community should devote greater efforts to addressing these disparities and collecting representative datasets.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Nowara et al. report insignificant differences in blood volume pulse SNR in iPPG between males and females.
“Nowara et al. report insignificant differences in blood volume pulse SNR in iPPG (imaging PPG) between males and females [112].”
From Sources of inaccuracy in photoplethysmography for continuous cardiovascular monitoring (2021) · cites this paper as [112]
Cites
The validity and practicality of sun-reactive skin types I through VI
Fitzpatrick, T. B. (1988) · Archives of Dermatology · doi:10.1001/archderm.1988.01670060015008 · Cited by 4,339 · see citing works
(2018) · Cited by 3,510 · see citing works
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., & Zemel, R. (2012) · Proceedings of the 3rd Innovations in Theoretical Computer Science Conference · doi:10.1145/2090236.2090255 · Cited by 3,501 · see citing works
Torralba, A., & Efros, A. A. (2011) · CVPR 2011 · doi:10.1109/cvpr.2011.5995347 · Cited by 2,440 · see citing works
MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition
Guo, Y., Zhang, L., Hu, Y., He, X., & Gao, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46487-9_6 · Cited by 2,122 · see citing works
Remote plethysmographic imaging using ambient light
Verkruysse, W., Svaasand, L. O., & Nelson, J. S. (2008) · Optics Express · doi:10.1364/oe.16.021434 · Cited by 1,728 · see citing works
A Multimodal Database for Affect Recognition and Implicit Tagging
Soleymani, M., Lichtenauer, J., Pun, T., & Pantic, M. (2012) · IEEE Transactions on Affective Computing · doi:10.1109/t-affc.2011.25 · Cited by 1,664 · see citing works
Non-contact, automated cardiac pulse measurements using video imaging and blind source separation
Poh, M., McDuff, D., & Picard, R. W. (2010) · Optics Express · doi:10.1364/oe.18.010762 · Cited by 1,652 · see citing works
Self-citationAdvancements in Noncontact, Multiparameter Physiological Measurements Using a Webcam
Poh, M., McDuff, D., & Picard, R. W. (2010) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2010.2086456 · Cited by 1,488 · see citing works
Self-citationRobust pulse rate from chrominance-based rPPG
de Haan, G., & Jeanne, V. (2013) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2013.2266196 · Cited by 1,378 · see citing works
Algorithmic principles of remote PPG
Wang, W., den Brinker, A. C., Stuijk, S., & de Haan, G. (2017) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2016.2609282 · Cited by 1,234 · see citing works
Umetani, K., Singer, D. H., McCraty, R., & Atkinson, M. (1998) · Journal of the American College of Cardiology · doi:10.1016/s0735-1097(97)00554-8 · Cited by 1,145 · see citing works
DeepPhys: Video-Based Physiological Measurement Using Convolutional Attention Networks
Chen, W., & McDuff, D. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01216-8_22 · Cited by 716 · see citing works
Self-citationDEX: Deep EXpectation of Apparent Age from a Single Image
Rothe, R., Timofte, R., & Gool, L. V. (2015) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2015.41 · Cited by 711 · see citing works
Pushing the frontiers of unconstrained face detection and recognition: IARPA Janus Benchmark A
Klare, B. F., Klein, B., Taborsky, E., Blanton, A., Cheney, J., Allen, K., Grother, P. J., Mah, A., Burge, M. J., & Jain, A. K. (2015) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2015.7298803 · Cited by 706 · see citing works
Raji, I. D., & Buolamwini, J. (2019) · Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3306618.3314244 · Cited by 651 · see citing works
The use of photoplethysmography for assessing hypertension
Elgendi, M., Fletcher, R., Liang, Y., Howard, N., Lovell, N. H., Abbott, D., Lim, K., & Ward, R. (2019) · npj Digital Medicine · doi:10.1038/s41746-019-0136-7 · Cited by 625 · see citing works
Multimodal Spontaneous Emotion Corpus for Human Behavior Analysis
Zhang, Z., Girard, J. M., Wu, Y., Zhang, X., Liu, P., Çiftçi, U. A., Canavan, S. J., Reale, M. J., Horowitz, A., Yang, H., et al. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.374 · Cited by 438 · see citing works
Non-contact video-based vital sign monitoring using ambient light and auto-regressive models
Tarassenko, L., Villarroel, M., Guazzi, A., Jorge, J., Clifton, D. A., & Pugh, C. W. (2014) · Physiological Measurement · doi:10.1088/0967-3334/35/5/807 · Cited by 421 · see citing works
A Novel Algorithm for Remote Photoplethysmography: Spatial Subspace Rotation
Wang, W., Stuijk, S., & de Haan, G. (2015) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2015.2508602 · Cited by 369 · see citing works
Yu, Z., Peng, W., Li, X., Hong, X., & Zhao, G. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00024 · Cited by 353 · see citing works
Heart rate measurement based on a time-lapse image.
Takano, C., & Ohta, Y. (2007) · Medical Engineering & Physics · doi:10.1016/j.medengphy.2006.09.006 · Cited by 345 · see citing works
Remote measurement of cognitive stress via heart rate variability
McDuff, D., Gontarek, S., & Picard, R. W. (2014) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2014.6944243 · Cited by 315 · see citing works
Self-citationVIPL-HR: A Multi-modal Database for Pulse Estimation from Less-Constrained Face Video
Niu, X., Han, H., Shan, S., & Chen, X. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-20873-8_36 · Cited by 232 · see citing works
A survey of remote optical photoplethysmographic imaging methods
McDuff, D., Estepp, J. R., Piasecki, A. M., & Blackford, E. B. (2015) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2015.7319857 · Cited by 212 · see citing works
Self-citationSparsePPG: Towards Driver Monitoring Using Camera-Based Vital Signs Estimation in Near-Infrared
Nowara, E. M., Marks, T. K., Mansour, H., & Veeraraghavan, A. (2018) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2018.00174 · Cited by 174 · see citing works
Self-citationInfluence of skin type and wavelength on light wave reflectance
Fallow, B. A., Tarumi, T., & Tanaka, H. (2013) · Journal of Clinical Monitoring and Computing · doi:10.1007/s10877-013-9436-7 · Cited by 162 · see citing works
Non-contact physiological monitoring of preterm infants in the Neonatal Intensive Care Unit
Villarroel, M., Chaichulee, S., Jorge, J., Davis, S., Green, G., Arteta, C., Zisserman, A., McCormick, K. P., Watkinson, P., & Tarassenko, L. (2019) · npj Digital Medicine · doi:10.1038/s41746-019-0199-5 · Cited by 155 · see citing works
Estepp, J. R., Blackford, E. B., & Meier, C. M. (2014) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2014.6974121 · Cited by 115 · see citing works
COGCAM: Contact-free Measurement of Cognitive Stress During Computer Tasks with a Digital Camera
McDuff, D. J., Hernandez, J., Gontarek, S., & Picard, R. W. (2016) · Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems · doi:10.1145/2858036.2858247 · Cited by 113 · see citing works
Self-citationBuolamwini, J. (2017) · DSpace@MIT (Massachusetts Institute of Technology) · Cited by 103 · see citing works
iPhys: An Open Non-Contact Imaging-Based Physiological Measurement Toolbox
McDuff, D., & Blackford, E. B. (2019) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2019.8857012 · Cited by 84 · see citing works
Self-citationSimultaneous monitoring of ballistocardiogram and photoplethysmogram using a camera
Shao, D., Tsow, F., Liu, C., Yang, Y., & Tao, N. (2017) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2016.2585109 · Cited by 69 · see citing works
Slope Transit Time (STT): A Pulse Transit Time Proxy requiring Only a Single Signal Fiducial Point
Addison, P. S. (2016) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2016.2528507 · Cited by 57 · see citing works
Characterizing Bias in Classifiers using Generative ModelsPreprint
McDuff, D., Ma, S., Song, Y., & Kapoor, A. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.11891 · Cited by 28 · see citing works
Self-citationImpact of makeup on remote-PPG monitoring
Wang, W., & Shan, C. (2019) · Biomedical Physics & Engineering Express · doi:10.1088/2057-1976/ab51ba · Cited by 27 · see citing works
Nowara, E. M., & McDuff, D. (2019) · doi:10.1109/iccvw.2019.00211 · Cited by 27 · see citing works
Self-citationAddison, P. S., Jacquel, D., Foo, D. M. H., & Borg, U. R. (2018) · Journal of Clinical Monitoring and Computing · doi:10.1007/s10877-017-0076-1 · Cited by 24 · see citing works
Improving Smiling Detection with Race and Gender Diversity
Ryu, H. J., Mitchell, M., & Adam, H. (2017) · Cited by 14 · see citing works
Nishidate, I., Tanabe, C., McDuff, D., Nakano, K., Niizeki, K., Aizu, Y., & Haneishi, H. (2019) · Optical Diagnostics and Sensing: Toward Point-of-Care Diagnostics · doi:10.1117/12.2509752 · Cited by 7 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Sources of inaccuracy in photoplethysmography for continuous cardiovascular monitoring
Fine, J., Branan, K. L., Rodriguez, A. J., Boonya-ananta, T., Ajmal, Ramella-Roman, J. C., McShane, M. J., & Coté, G. L. (2021) · Biosensors · doi:10.3390/bios11040126 · Cited by 348 · see citing works
Camera Measurement of Physiological Vital Signs
McDuff, D. (2022) · ACM Computing Surveys · doi:10.1145/3558518 · Cited by 167 · see citing works
“Some initial work has begun to characterize diferences in performance of algorithms (both supervised and unsupervised) by diferent demographic and environmental parameters [3, 107].”
BackgroundSelf-citationDeep Learning Methods for Remote Heart Rate Measurement: A Review and Future Research Agenda
Cheng, C.-H., Wong, K. L., Chin, J. W., Chan, T. T., & So, R. H. Y. (2021) · Sensors · doi:10.3390/s21186296 · Cited by 122 · see citing works
“Other challenges, such as skin-tone variations [22,160,161], multiple persons detection [14,122], and long distance estimations [14,162], need to be overcome, as well, if the methods are to be, ultimately, robust and highly applicable in the real-world, replacing all contact-based methods.”
BackgroundMMPD: Multi-Domain Mobile Video Physiology Dataset
Tang, J., Chen, K., Wang, Y., Shi, Y., Patel, S., McDuff, D., & Liu, X. (2023) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc40787.2023.10340857 · Cited by 75 · see citing works
“Research has shown that it is harder to extract pulse signals from individuals with darker skin tones due to the lower signal-to-noise ratio in the reflected light [3].”
BackgroundSelf-citationSun, Z., & Li, X. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3367910 · Cited by 71 · see citing works
“Previous studies [67], [79], [80] have pointed out that rPPG algorithms based on RGB videos have performance bias in dif-ferentskintones.”
BackgroundPhotoplethysmography—new applications for an old technology: a sleep technology review
Ryals, S. M., Chiang, A. A.-P., Schutte-Rodin, S., Chandrakantan, A., Verma, N., Holfinger, S. J., Abbasi-Feinberg, F., Bandyopadhyay, A., Baron, K. G., Bhargava, S., et al. (2022) · Journal of Clinical Sleep Medicine · doi:10.5664/jcsm.10300 · Cited by 43 · see citing works
Remote photoplethysmography (rPPG) in the wild: Remote heart rate imaging via online webcams
di Lernia, D., Finotti, G., Tsakiris, M., Riva, G., & Naber, M. (2024) · Behavior Research Methods · doi:10.3758/s13428-024-02398-0 · Cited by 43 · see citing works
“Moreover, it must be noted that previous studies showed that rPPG performance drops significantly with darker skin tones (for a review, see Nowara et al. (2020)).”
BackgroundBlood Pressure Measurement: From Cuff-Based to Contactless Monitoring
Man, P.-K., Cheung, K.-L., Sangsiri, N., Shek, W., Wong, K. L., Chin, J. W., Chan, T. T., & So, R. H. Y. (2022) · Healthcare · doi:10.3390/healthcare10102113 · Cited by 40 · see citing works
“However, remote BP prediction also has susceptibilities in handling noise and artifacts unrelated to the hemoglobin signals, such as makeup [203], skin tone [204,205], illumination [206,207], camera distance [208,209], camera specification [210,211], and subject motion [212,213].”
BackgroundPremkumar, S., & Hemanth, D. J. (2022) · Informatics · doi:10.3390/informatics9030057 · Cited by 37 · see citing works
“Subsequently, the authors of [92] proposed a deep super-resolution network for lowresolution video which enhanced the rPPG method in compressed video and conducted a performance analysis at varying compression levels and in different formats.”
MethodContactless Blood Oxygen Saturation Estimation from Facial Videos Using Deep Learning
Cheng, C.-H., Yuen, Z., Chen, S., Wong, K. L., Chin, J. W., Chan, T. T., & So, R. H. Y. (2024) · Bioengineering · doi:10.3390/bioengineering11030251 · Cited by 36 · see citing works
A Survey of Photoplethysmography and Imaging Photoplethysmography Quality Assessment Methods
Desquins, T., Bousefsaf, F., Pruski, A., & Maaoui, C. (2022) · Applied Sciences · doi:10.3390/app12199582 · Cited by 33 · see citing works
Ouzar, Y., Djeldjli, D., Bousefsaf, F., & Maaoui, C. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.106592 · Cited by 31 · see citing works
Haugg, F., Elgendi, M., & Menon, C. (2023) · Bioengineering · doi:10.3390/bioengineering10020243 · Cited by 31 · see citing works
“The literature almost unanimously ascertains that darker skin tends to make rPPG methods more error-prone [31,32].”
Contactless Vital Sign Monitoring: A Review Towards Multi-Modal Multi-Task Approaches
Hassanpour, A., & Yang, B. (2025) · Sensors · doi:10.3390/s25154792 · Cited by 28 · see citing works
Using explainable machine learning and fitbit data to investigate predictors of adolescent obesity
Kiss, O., Baker, F. C., Pálovics, R., Dooley, E. E., Gabriel, K. P., & Nagata, J. M. (2024) · Scientific Reports · doi:10.1038/s41598-024-60811-2 · Cited by 22 · see citing works
Chapman, B. P., Gullapalli, B. T., Rahman, T., Smelson, D. A., Boyer, E. W., & Carreiro, S. (2022) · npj Digital Medicine · doi:10.1038/s41746-022-00664-z · Cited by 21 · see citing works
Robust in-vehicle heartbeat detection using multimodal signal fusion
Warnecke, J. M., Lasenby, J., & Deserno, T. M. (2023) · Scientific Reports · doi:10.1038/s41598-023-47484-z · Cited by 18 · see citing works
“This is 3.2% higher than with the sensor-based late fusion with ECG+PPG+iPPG.”
ResultInfluentialSensor Fusion for Robust Heartbeat Detection during Driving
Warnecke, J. M., Boeker, N., Spicher, N., Wang, J., Flormann, M., & Deserno, T. M. (2021) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc46164.2021.9630935 · Cited by 17 · see citing works
“Besides, the skin type and melanin level have an impact on the SNR and the iPPG analysis [20].”
BackgroundEffects of illuminance intensity on the green channel of remote photoplethysmography (rPPG) signals
Guler, S., Öztürk, Ö., Golparvar, A. J., Doğan, H., & Yapici, M. K. (2022) · Physical and Engineering Sciences in Medicine · doi:10.1007/s13246-022-01175-7 · Cited by 16 · see citing works
For Heart Rate Assessments from Drone Footage in Disaster Scenarios
Mösch, L., Barz, I., Müller, A., Pereira, C. B., Moormann, D., Czaplik, M., & Follmann, A. (2023) · Bioengineering · doi:10.3390/bioengineering10030336 · Cited by 12 · see citing works
“Individuals with higher melanin content in their skin, which absorbs a greater proportion of environmental light than those with lower melanin content, tend to have less accurate rPPG measurements [25,37].”
Background“Warm Bodies”: A Post-Processing Technique for Animating Dynamic Blood Flow on Photos and Avatars
McDuff, D., & Nowara, E. M. (2021) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445719 · Cited by 12 · see citing works
“which absorbs more light inside the skin [34].”
BackgroundSelf-citationMetaPhys: Few-Shot Adaptation for Non-Contact Physiological MeasurementPreprint
Liu, X., Jiang, Z., Fromm, J., Xu, X., Patel, S., & McDuff, D. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.01773 · Cited by 12 · see citing works
“One specific example is that there exists biases in performance across skin types Nowara et al. (2020).”
BackgroundSelf-citationA survey of blood oxygen saturation assessment from video
Wuyart, A., Vuillaume, L. A., Maaoui, C., & Bousefsaf, F. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.108069 · Cited by 11 · see citing works
Demographic bias in public remote photoplethysmography datasets
Bondarenko, M., Menon, C., & Elgendi, M. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01973-9 · Cited by 11 · see citing works
“Therefore, in order to study the factors that affect the accuracy of HR estimation, numerous scholars have investigated several aspects.”
Dynamic Region of Interest Selection in Remote Photoplethysmography: Proof-of-Concept Study
Kiddle, A., Barham, H., Wegerif, S., & Petronzio, C. (2023) · JMIR Formative Research · doi:10.2196/44575 · Cited by 10 · see citing works
Adams, A. T., Mandel, I., Gao, Y., Heckman, B. W., Nandakumar, R., & Choudhury, T. (2022) · JMIR Biomedical Engineering · doi:10.2196/34934 · Cited by 10 · see citing works
Raw Photoplethysmography as an Enhancement for Research-Grade Wearable Activity Monitors
Hibbing, P. R., & Khan, M. M. (2024) · JMIR mhealth and uhealth · doi:10.2196/57158 · Cited by 10 · see citing works
Race Correction and Algorithmic Bias in Atrial Fibrillation Wearable Technologies
Merid, B., & Volpe, V. V. (2023) · Health Equity · doi:10.1089/heq.2023.0034 · Cited by 9 · see citing works
“40,41 Most PPG devices use green light due to the high absorption spectrum of hemoglobin, yet, as we note earlier, the absorption of green light by melanin for those with darker skin limits light penetration and gives inaccurate readings.”
BackgroundA Framework for Selecting and Assessing Wearable Sensors Deployed in Safety Critical Scenarios
Houghton, R., Martinetti, A., & Majumdar, A. (2024) · Sensors · doi:10.3390/s24144589 · Cited by 8 · see citing works
Diverse R-PPG: Camera-Based Heart Rate Estimation for Diverse Subject Skin-Tones and ScenesPreprint
Chari, P., Kabra, K., Karinca, D., Lahiri, S., Srivastava, D., Kulkarni, K., Chen, T., Cannesson, M. P., Jalilian, L., & Kadambi, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.12769 · Cited by 12 · see citing works
Biometric Authentication Based on Enhanced Remote Photoplethysmography Signal Morphology
Sun, Z., Li, X., Komulainen, J., & Zhao, G. (2024) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb62174.2024.10744426 · Cited by 7 · see citing works
“Regarding fairness, prior studies [30, 42] highlighted skin bias in rPPG signal quality.”
BackgroundInfluential Factors in rPPG: Insights from a Diverse and Inclusive Empirical Study
Rivest‐Hénault, D., Proulx, C., Ho, K., Jafari, N., Yakubu, T., Torres, S., Lim, M., Pecora, L., Bernhardt, R., Auer, S., et al. (2024) · Wearable and Implantable Body Sensor Networks · doi:10.1109/bsn63547.2024.10780670 · Cited by 5 · see citing works
Cramer, I. C., van Esch, R., Verstappen, C. C. A. G., Kloeze, C., van Bussel, B. C. T., Stuijk, S., Bergmans, J. W. M., Veer, M. V. ’., Zinger, S., Montenij, L. J., et al. (2025) · Journal of Clinical Monitoring and Computing · doi:10.1007/s10877-025-01263-5 · Cited by 6 · see citing works
Combining Magnification and Measurement for Non-Contact Cardiac Monitoring
Nowara, E. M., McDuff, D., & Veeraraghavan, A. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00422 · Cited by 6 · see citing works
“Hence, these methods often struggle with generalizing to new data that may have different motion [2, 37], different video compression [6], or even participants with different skin types or genders [15].”
BackgroundInfluentialSelf-citationArrow, C., Ward, M., Eshraghian, J. K., & Dwivedi, G. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110956 · Cited by 5 · see citing works
Overcoming Difficulty in Obtaining Dark-skinned Subjects for Remote-PPG by Synthetic AugmentationPreprint
Ba, Y., Wang, Z., Karinca, D., Bozkurt, O. D., & Kadambi, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.06007 · Cited by 10 · see citing works
“With the training sets heavily biased towards subjects of light skin tones, the state-of-the-art data-driven rPPG models usually fail to generalize their performance to the underrepresented groups [22].”
BackgroundPhysFlow: Skin tone transfer for remote heart rate estimation through conditional normalizing flows
Comas, J., Alomar, A., Ruiz, A., & Sukno, F. (2024) · arXiv · doi:10.48550/arxiv.2407.21519 · Cited by 4 · see citing works
“Firstly, across all methods, there is a noticeable decline in heart rate performance with darker skin tones, consistent with findings in previous studies [2, 15, 38].”
BackgroundMethodResultInfluentialLSTS: Periodicity Learning via Long Short-Term Temporal Shift for Remote Physiological Measurement
Jiang, T., Ma, Y., Li, J., Dong, Q., Ji, X., & Liu, Y. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3538474 · Cited by 4 · see citing works
Pstras, L., Okupnik, T., Ponikowska, B., & Paleczny, B. (2025) · Frontiers in Physiology · doi:10.3389/fphys.2025.1638809 · Cited by 4 · see citing works
Williams, J. T., Francombe, J., & Murphy, D. T. (2023) · Applied Sciences · doi:10.3390/app13148482 · Cited by 4 · see citing works
MIME: Minority Inclusion for Majority Group Enhancement of AI Performance
Chari, P., Ba, Y., Athreya, S., & Kadambi, A. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19778-9_19 · Cited by 3 · see citing works
Marmuse, A., Billaud, J.-B., Jacob, S., Vigier, C., Ramdani, C., & Trousselard, M. (2024) · Military Psychology · doi:10.1080/08995605.2024.2324645 · Cited by 4 · see citing works
“Non-contact photo-plethysmography can be performed on all skin types under certain skin color conditions, although mean performance has been observed to fall significantly for the darkest skin phototype (Nowara et al., 2020).”
Moustafa, M., Farooq, M. A., Elrasad, A., Lemley, J., & Corcoran, P. (2024) · IEEE Access · doi:10.1109/access.2024.3472508 · Cited by 3 · see citing works
“Meanwhile, other factors, such as the heart rate ranges in the dataset [38] and subject diversity (gender, ethnicity, etc.) [39], can affect how well a model generalizes to other datasets.”
BackgroundImplicit Neural Models to Extract Heart Rate from Video
Chari, P., Harish, A. B., Armouti, A., Vilesov, A., Sarda, S., Jalilian, L., & Kadambi, A. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73010-8_10 · Cited by 3 · see citing works
Woelk, S. P., Garfinkel, S. N., Mayiwar, L., & Knöferle, K. (2026) · Behavior Research Methods · doi:10.3758/s13428-026-02953-x · Cited by 2 · see citing works
Dasa, D., & Davies, P. (2026) · BMJ Open · doi:10.1136/bmjopen-2026-119311 · Cited by 2 · see citing works
Smart Wristband Based on a Magnetic Soft Sensor for Pulse Wave Measurement
Romano, C., Silvestri, S., Schena, E., & Massaroni, C. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3594238 · Cited by 2 · see citing works
“However, despite its convenience, PPG technology suffers from several limitations, including susceptibility to light intensity fluctuations, skin tone variability, and sensor-to-artery distance [15], [16].”
Robust In-Vehicle Heartbeat Detection Using Multimodal Signal FusionPreprint
Warnecke, J. M., Lasenby, J., & Deserno, T. M. (2023) · Research Square · doi:10.21203/rs.3.rs-2828298/v1 · Cited by 2 · see citing works
Camera-Sourced Heart Rate Synchronicity: A Measure of Immersion in Audiovisual Experiences
Williams, J., Francombe, J., & Murphy, D. T. (2024) · Applied Sciences · doi:10.3390/app14167228 · Cited by 1 · see citing works
Seo, W.-D., Choi, I., Kim, G. C., Song, H. S., & Kim, Y. K. (2026) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2026.109852 · Cited by 1 · see citing works
Technologies of antiblackness: Black matter, racial bias, and the haunting past of the spirometer
Barla, J. (2023) · Technology in Society · doi:10.1016/j.techsoc.2023.102256 · Cited by 3 · see citing works
Dasa, D., & Davies, P. (2025) · medRxiv · doi:10.1101/2025.10.13.25337878 · Cited by 1 · see citing works
“However, training/validation datasets frequently under-represent darker skin, risking performance disparities [16, 17, 18].”
InfluentialTowards Collecting Big Data for Remote Photoplethysmography
Kalinin, K. B., Mironenko, Y., Kopeliovich, M. V., & Petrushan, M. V. (2021) · Lecture notes in networks and systems · doi:10.1007/978-3-030-80129-8_6
Racial Deviations in Smartphone and Wearable Sensor Measurements of Health BiometricsPreprint
Lam, E., Marais, P., & Leung, H. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4663485
Kawasaki, S., Ashida, K., Nguyen, V.-T., Ngo, T. D., Le, D.-D., Doi, H., & Tsumura, N. (2024) · Applied Sciences · doi:10.3390/app14156697
Sun, J., Li, K., Li, Z., Zhang, M., Wu, Z., & Zhang, D. N. (2024) · Journal of Biophotonics · doi:10.1002/jbio.202400448
ML and DL Model Comparison with Skin Tone and Dynamic Range Analysis for rPPG
Jayasinghe, J. A. S. Y., & Malasinghe, L. (2024) · International Conference on Advancements in Computing (ICAC) · doi:10.1109/icac64487.2024.10850984
Pstras, L., Okupnik, T., Ponikowska, B., & Paleczny, B. (2025) · medRxiv · doi:10.1101/2025.02.13.25322005
Investigating the Effectiveness of Filter in rPPG Methods Across Various Skin Tones
ASLAMI, F., & Sripian, P. (2025) · International Journal of Affective Engineering · doi:10.5057/ijae.ijae-d-24-00035
Designing Reproducible Test Environments for rPPG: A System for Camera Sensor Response Validation
van Putten, L. D., Veleslavov, I., Ahmed, A., Mathieu, A., & Wegerif, S. (2026) · Lights · doi:10.3390/lights2020003
“These make the response largely independent of illumination intensities similar to the processing steps commonly used in rPPG processing, such as AC/DC normalisation [29,30] or by subtracting the pixel norm and dividing by the pixel standard deviation [31].”
Ying, C., Yang, H., 葛介艺, Sun, Z., Cheng, X., Ren, K., & Li, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00874
“While visible light contains rich physiological information, its reliability can deteriorate under specific circumstances [27, 28] , such as when monitoring subjects with darker skin tones or in environments with dramatic lighting variations.”
Reducing Annotation Bias and Burden in rPPG Research Using Automated Monk Skin Tone Classification
Alhasani, T., Alhameli, M., Ahmed, S. G., Alnajjar, F., & Jassmi, H. A. (2026) · Student Research Conference on Applied Computing (SRC) · doi:10.1109/src70627.2026.11550603
“Still, recent work calls for an assessment of rPPG systems under different skin tones and controlled lighting conditions, instead of aggregating the rPPG results across subjects [5].”
InfluentialFinotti, G., Lernia, D. D., & Manos Tsakiris, A. (2026) · Behavior Research Methods · doi:10.3758/s13428-026-03098-7
Moon, S. I., & Kwon, N. K. (2026) · Electronics · doi:10.3390/electronics15153477
Equitable rPPG: Heart Rate Estimation for High-Melanin Skin using Triple-Stream 3D-CNNs
Ahamad, M. M., Alam, K. S., Aktar, S., Hubbard, J., & Ahamed, S. I. (2026) · IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies · doi:10.1109/chase69719.2026.00032
“The model showed a steady increase in MAE with increasing Fitzpatrick scale, consistent with melanin’s optical absorption properties [15].”
InfluentialAge and Skin Tone Biases in Remote Heart Rate Monitoring
Ma, B., Kaetheeswaran, J., Abedi, A., Lankarany, M., & Khan, S. S. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-5190-2_10
Estimating heart rate variability using facial video photoplethysmography: a pilot validation study
Pstras, L., Okupnik, T., Ponikowska, B., & Paleczny, B. (2026) · Scientific Reports · doi:10.1038/s41598-026-70292-0
“Second, we studied only white subjects, in whom vPPG technology is known to perform better compared to dark-skinned individuals 64,65 .”
- Authors
- Nowara, E. M., McDuff, D., & Veeraraghavan, A.
- Affiliations
- Rice University (US)
- Microsoft (United States) (US)
- Venue
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
- Publisher
- IEEE
- Published
- 2020-06
- Pages
- 1148-1155
- OpenAlex
- W3036711870
- Cited by
- 66
- References
- 40
- Citations by year
- 2026: 10 · 2025: 12 · 2024: 15 · 2023: 11 · 2022: 9 · 2021: 7 · 2020: 2
- FWCI
- 0.61
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (publishedVersion)
- Topics
- Physical Sciences › Engineering › Biomedical Engineering › Non-Invasive Vital Sign Monitoring
- Physical Sciences › Computer Science › Signal Processing › Biometric Identification and Security
- Health Sciences › Medicine › Cardiology and Cardiovascular Medicine › Heart Rate Variability and Autonomic Control
- Keywords
- imaging photoplethysmography, skin type, gender bias, heart rate estimation, non-contact vital sign monitoring, dataset representativeness, algorithmic fairness, physiological monitoring
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
