pith. sign in

arxiv: 1906.08192 · v1 · pith:4RX6VKPJnew · submitted 2019-06-19 · 📡 eess.IV · eess.SP

Exploiting Weak Head Movements for Camera-based Respiration Detection

Fabian Schrumpf , Christoph Moench , Gerold Bausch , Mirco Fuchs This is my paper

Pith reviewed 2026-05-25 20:05 UTC · model grok-4.3

classification 📡 eess.IV eess.SP
keywords respiration rate estimationhead movement detectionfacial video analysisnon-contact monitoringrPPG alternativechest movement couplingcamera-based vital signs
0
0 comments X

The pith

Weak head movements induced by breathing produce detectable signals in facial video for respiration rate estimation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tests whether mechanical coupling from chest breathing creates minor head motions that can be picked up as subtle intensity changes in face video, using both motion tracking and color analysis. This matters in camera setups that show only the head, where chest regions are out of frame and skin-based color signals often fail. An experiment with 12 subjects extracted respiratory frequency estimates from head regions and compared them to chest-based estimates. The results indicate that correlated signals can be obtained from the face. Such an approach would extend non-contact monitoring to more restricted viewing angles without added hardware.

Core claim

The mechanical coupling between chest and head induces minor movements of the head which, like in rPPG, can be detected from subtle color changes as well. Although the strength of these movements is expected to be much smaller in scale, sensing these intensity variations could provide a reasonably suited respiration signal for subsequent respiratory rate analysis. Results from 12 subjects show that it is possible to derive signals correlated to chest movement from facial regions, offering a feasible alternative when rPPG-signals cannot be derived reliably and chest movement detection cannot be applied.

What carries the argument

Application of motion- and rPPG-based methods to facial regions to capture breathing-linked head movements as intensity variations.

If this is right

  • Respiratory signals can be extracted from head regions in close-up head imaging where chest is invisible.
  • The approach serves as backup when rPPG fails due to lack of visible skin.
  • Respiratory frequency estimates from head and chest regions can be obtained with the same processing pipeline.
  • The method requires only standard video and no contact sensors.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The technique could support respiration tracking during video calls or driver monitoring where only the face is framed.
  • Fusing head-derived signals with occasional chest glimpses might increase robustness across lighting changes.
  • Further tests could check performance when subjects move their heads voluntarily or under varying illumination.

Load-bearing premise

Breathing produces head movements that are large enough to create measurable color intensity changes in video of the face.

What would settle it

A controlled recording where chest and head video are captured simultaneously but head-region signals show no statistical correlation with measured respiration rate or chest signals.

Figures

Figures reproduced from arXiv: 1906.08192 by Christoph Moench, Fabian Schrumpf, Gerold Bausch, Mirco Fuchs.

Figure 1
Figure 1. Figure 1: Experimental protocol. The respiratory frequency was varied [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 4
Figure 4. Figure 4: SNR of the respiratory signal derived from the RGB and rPPG [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Color mapping of the SNR-values of every sub-ROI. Background areas were not considered for SNR estimation. [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
read the original abstract

In recent years, considerable progress has been made in the non-contact based detection of the respiration rate from video sequences. Common techniques either directly assess the movement of the chest due to breathing or are based on analyzing subtle color changes that occur as a result of hemodynamic properties of the skin tissue by means of remote photoplethysmography (rPPG). However, extracting hemodynamic parameters from rPPG is often difficult especially if the skin is not visible to the camera. In contrast, extracting respiratory signals from chest movements turned out to be a robust method. However, the detectability of chest regions cannot be guaranteed in any application scenario, for instance if the camera setting is optimized to provide close-up images of the head. In such a case an alternative method for respiration detection is required. It is reasonable to assume that the mechanical coupling between chest and head induces minor movements of the head which, like in rPPG, can be detected from subtle color changes as well. Although the strength of these movements is expected to be much smaller in scale, sensing these intensity variations could provide a reasonably suited respiration signal for subsequent respiratory rate analysis. In order to investigate this coupling we conducted an experimental study with 12 subjects and applied motion- and rPPG-based methods to estimate the respiratory frequency from both head regions and chest. Our results show that it is possible to derive signals correlated to chest movement from facial regions. The method is a feasible alternative to rPPG-based respiratory rate estimations when rPPG-signals cannot be derived reliably and chest movement detection cannot be applied as well.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript reports an experimental study with 12 subjects showing that respiratory frequency can be estimated from facial video regions by applying motion- and rPPG-based methods to exploit weak head movements mechanically coupled to chest respiration. This is positioned as a feasible alternative when chest regions are not visible to the camera and standard rPPG signals cannot be derived reliably.

Significance. If validated, the approach extends non-contact respiration monitoring to head-only camera views, complementing existing rPPG and chest-motion techniques. The small cohort and absence of detailed statistical reporting, however, constrain the strength of the feasibility claim and its immediate applicability.

major comments (2)
  1. [Abstract/Results] Abstract and Results: positive outcomes are reported from the 12-subject study, yet no information is supplied on statistical tests performed, error metrics, subject exclusion criteria, or controls for voluntary head movements; these omissions directly affect the verifiability of the central feasibility claim.
  2. [Methods] Methods: the description of how motion- and rPPG-based estimators were applied to head versus chest regions lacks quantitative details on signal processing parameters, region-of-interest selection, or handling of low-amplitude head motion, making it difficult to assess whether the detected correlations arise from the hypothesized mechanical coupling.

Simulated Author's Rebuttal

2 responses · 0 unresolved

Thank you for the opportunity to respond to the referee's report. We address each major comment below and indicate the revisions made to the manuscript.

read point-by-point responses

  1. Referee: [Abstract/Results] Abstract and Results: positive outcomes are reported from the 12-subject study, yet no information is supplied on statistical tests performed, error metrics, subject exclusion criteria, or controls for voluntary head movements; these omissions directly affect the verifiability of the central feasibility claim.

    Authors: We agree that these details were omitted from the original submission. In the revised manuscript we have added the statistical tests performed (Pearson correlation and paired comparisons with p-values), error metrics (mean absolute error and mean absolute percentage error on estimated respiratory frequencies), confirmation that no subjects were excluded, and a description of the protocol instructions given to participants to remain still and avoid voluntary head movements. These additions appear in the Methods, Results, and updated Abstract. revision: yes

  2. Referee: [Methods] Methods: the description of how motion- and rPPG-based estimators were applied to head versus chest regions lacks quantitative details on signal processing parameters, region-of-interest selection, or handling of low-amplitude head motion, making it difficult to assess whether the detected correlations arise from the hypothesized mechanical coupling.

    Authors: We concur that additional quantitative information is required. The revised Methods section now specifies the exact filter cut-offs, window lengths, and sampling parameters used for both the motion-based and rPPG-based pipelines; describes the ROI selection procedure (facial landmark-based regions for the head and bounding-box regions for the chest); and explains the handling of low-amplitude motion via the chosen optical-flow and rPPG algorithms. These details enable readers to evaluate the mechanical-coupling hypothesis. revision: yes

Circularity Check

0 steps flagged

No significant circularity: experimental feasibility study with direct empirical comparison

full rationale

The paper reports an experimental study on 12 subjects that applies existing motion- and rPPG-based respiratory frequency estimators to head versus chest regions and compares the resulting signals. No equations, parameter fits, derivations, or self-citation chains are present in the provided text or abstract. The central claim (feasibility of deriving correlated signals from facial regions) rests on direct measurement against chest reference, not on any reduction to inputs by construction. This is the most common honest non-finding for purely empirical papers.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The claim rests on the domain assumption of detectable mechanical coupling and standard video signal processing assumptions; no free parameters or invented entities are introduced.

axioms (1)
  • domain assumption Mechanical coupling between chest and head produces detectable minor head movements from breathing.
    Explicitly stated as a reasonable assumption in the abstract.

pith-pipeline@v0.9.0 · 5824 in / 1098 out tokens · 25521 ms · 2026-05-25T20:05:25.746105+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

  1. [1]

    Impedance pneumography: Comparison between chest impedance changes and respiratory volumes in 11 healthy volunteers,

    A. Grenvik, S. Ballou, E. McGinley, J. E. Millen, W. L. Cooley, and P. Safar, “Impedance pneumography: Comparison between chest impedance changes and respiratory volumes in 11 healthy volunteers,” Chest Journal, vol. 62, no. 4, pp. 439–443, 1972

    work page 1972

  2. [2]

    An assessment of algorithms to estimate res- piratory rate from the electrocardiogram and photoplethysmogram

    P. H. Charlton, T. Bonnici, L. Tarassenko, D. A. Clifton, R. Beale, and P. J. Watkinson, “An assessment of algorithms to estimate res- piratory rate from the electrocardiogram and photoplethysmogram.” Physiological measurement, vol. 37, no. 4, pp. 610–26, 2016

    work page 2016

  3. [3]

    A new look at the essence of the imaging photoplethysmography

    A. A. Kamshilin et al. , “A new look at the essence of the imaging photoplethysmography.” Scientific reports, vol. 5, p. 10494, 2015

    work page 2015

  4. [4]

    Remote plethysmo- graphic imaging using ambient light

    W. Verkruysse, L. O. Svaasand, and J. S. Nelson, “Remote plethysmo- graphic imaging using ambient light.” Optics express, vol. 16, no. 26, pp. 21 434–45, 2008

    work page 2008

  5. [5]

    Robust respiration detec- tion from remote photoplethysmography,

    M. van Gastel, S. Stuijk, and G. de Haan, “Robust respiration detec- tion from remote photoplethysmography,” Biomedical Optics Express, vol. 7, no. 12, p. 4941, 2016

    work page 2016

  6. [6]

    Photoplethysmography and its application in clinical physio- logical measurement,

    J. Allen, “Photoplethysmography and its application in clinical physio- logical measurement,” Phys. Meas., vol. 28, no. 3, pp. R1–R39, 2007

    work page 2007

  7. [7]

    Remote monitoring of camera based respiration rate estimated by using occlusion of dot pattern,

    V . V . Makkapati and S. S. Rambhatla, “Remote monitoring of camera based respiration rate estimated by using occlusion of dot pattern,” 2016 IEEE Int. Conf. Advanced Netw. Telecommun. Syst. , pp. 1–5, 2017

    work page 2016

  8. [8]

    Camera-based system for contactless monitoring of respiration,

    M. Bartula, T. Tigges, and J. Muehlsteff, “Camera-based system for contactless monitoring of respiration,” Proce. Annu. Int. Conf. IEEE Eng. in Med. and Biol. Soc., EMBS , pp. 2672–2675, 2013

    work page 2013

  9. [9]

    The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis,

    N. E. Huang et al., “The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis,” Royal Soc. of London Proc. , vol. 454, no. 1, pp. 903–995, 1996

    work page 1996

  10. [10]

    Remote heart rate measurement from face videos under realistic situations,

    X. Li, J. Chen, G. Zhao, and M. Pietikinen, “Remote heart rate measurement from face videos under realistic situations,” Proc. IEEE Comput. Soc. Conf. Comput. Vision and Pattern Recognition , pp. 4264–4271, 2014

    work page 2014

  11. [11]

    Estimation of respiration rate from ECG, BP and PPG signals using empirical mode decomposition,

    K. V . Madhav, M. R. Ram, E. H. Krishna, N. R. Komalla, and K. A. Reddy, “Estimation of respiration rate from ECG, BP and PPG signals using empirical mode decomposition,” in IEEE Instrum. Meas. Technol. Conf. IEEE, 2011, pp. 1661–1664

    work page 2011

  12. [12]

    Monitoring of respiratory and heart rates using a fibre-optic sensor,

    L. G. Lindberg, H. Ugnell, and P. ˚A. berg, “Monitoring of respiratory and heart rates using a fibre-optic sensor,” Medical & Biological Engineering & Computing , vol. 30, no. 5, pp. 533–537, 1992

    work page 1992

  13. [13]

    Derivation of the respiratory rate from directly and indirectly measured respiratory signals using autocorrelation,

    F. Schrumpf, M. Sturm, G. Bausch, and M. Fuchs, “Derivation of the respiratory rate from directly and indirectly measured respiratory signals using autocorrelation,” Curr. Dir. Biomed. Eng., vol. 2, no. 1, pp. 241–245, 2016

    work page 2016

  14. [14]

    Relation between pulse pressure and the pulsation strength in camera-based photoplethysmograms,

    A. Trumpp et al., “Relation between pulse pressure and the pulsation strength in camera-based photoplethysmograms,” Current Directions in Biomedical Engineering , vol. 3, no. 2, 2017

    work page 2017