Skip to content

Accuracy & Guidelines

TL;DR: VitalLens achieves high fidelity on high-quality video inputs, but input quality matters.

In our study conducted on a diverse benchmark of 422 unique individuals, we observed a mean absolute error of:

  • Heart Rate: 1.57 bpm
  • Respiratory Rate: 1.08 breaths/min
  • HRV-SDNN: 10.18 ms

Guidelines for Best Accuracy

The "Garbage In, Garbage Out" rule applies strictly to rPPG. To maximize the confidence scores returned by the API, adhere to these guidelines:

1. Lighting Environment

  • Even Illumination: Ensure the face is evenly lit. Shadows moving across the face will create artifacts.
  • Avoid Flickering: Older fluorescent tubes and some LED lights flicker at frequencies that interfere with pulse detection.
  • Monitor Glare: Avoid having the subject sit in a dark room illuminated only by a computer monitor. Screens refresh at specific rates and emit changing colors that can corrupt the rPPG signal on the subject's face.

2. Video Quality

  • Compression: Avoid heavy video compression (low bitrate). rPPG relies on imperceptible skin color changes. Compression algorithms are designed to remove exactly these "redundant" color details.
  • Resolution: While the API downsamples images internally, inputting uncompressed or high-bitrate video (even at 480p) is preferred over 4K video with high compression artifacts.

3. Subject Positioning

  • Stability: Use a tripod or a stable surface. Handheld vibration introduces motion noise that obscures the rPPG signal.
  • Orientation: The subject should face the camera directly.
  • Framing: Include the upper chest in the frame.
    • Face Only: Good for Heart Rate.
    • Face + Upper Chest: Required for accurate Respiratory Rate, as the model analyzes subtle torso motion cues.

For a deeper technical analysis of how environmental factors impact signal integrity, read our technical paper.