GlowPulse
Your Mac's camera is now a heart-rate sensor
Start new thread

I tested GlowPulse against a Polar H10 chest strap – here are the real numbers

by•

At launch a few of you (rightly) asked how accurate reading heart rate from a webcam actually is. So I stopped hand-waving and measured it.

Setup
GlowPulse and a Polar H10 (chest-strap ECG, the practical gold standard) recording at the same time, second by second, across 9 sessions and ~3.5 hours total, in good light, dim light, movement, and dark.

What I found

  1. Good light, sitting still: off by about 1 BPM from the strap on average (0.6 to 1.3 across sessions), within 5 BPM 95% of the time. Shows a value the whole time.

  2. Dim light / movement: error grows to roughly 2.5 to 2.9 BPM. It follows the trend and only lags the fastest spikes.

  3. Dark: it mostly goes quiet, showing a number about 1 time in 10. I'd rather show nothing than something wrong.

To be clear
GlowPulse gives wellness estimates, not medical measurements. Polar H10 is named only as a neutral reference, no endorsement implied.

I published the full method, the per-session numbers, the analysis script, and the raw data so anyone can check it:

Happy to get into the rPPG pipeline or how I ran the comparison.

36 views

Add a comment

Replies

Best

I really like that you published the raw data and methodology instead of just saying "it's accurate."


For products like this, transparency builds a lot of trust. Also, I think showing no reading in poor conditions is much better than showing an inaccurate one just to keep numbers on the screen.

Curious, have you tested it across different webcams too? I wonder how much camera quality affects the results.

 Thanks Hossein, wasn't really a tough call for me, I just copied Apple's own approach: the Watch shows a "–" instead of a number when it can't get a clean signal, rather than guessing. Honest silence beats a confident wrong number.

On webcams: most of my testing is on the built-in Mac camera. Resolution barely matters, the signal is a tiny color change averaged over a skin patch, so megapixels don't help. What actually causes drift is the camera's auto-exposure and white-balance constantly readjusting, plus lighting. One user did hit drift on an external USB cam, which pushed a v1.1 fix.

Really appreciate seeing the raw comparison instead of just a vague “it’s accurate” claim. The part I like most is that GlowPulse goes quiet in bad conditions rather than forcing a number on the screen. For a wellness tool, that kind of honesty actually builds more trust.

 Thanks!