Ventures
Signal layer / API · API / Developer · Founder
VitalSign AI
Contactless pulse-rate estimation from an ordinary webcam, delivered as an API.
vitalsign.aiPosition in the stack
The input layer of the stack. If an interface is going to adapt to a person, something has to measure the person — without a wearable, and without asking them to do anything.
Somatic AI angle — The scalable physiological input layer every other venture reads from.
What it is
- An rPPG API: pulse-rate estimation from standard camera input, no additional hardware.
- Additional experimental outputs described by the venture — breathing rate, posture, drowsiness, blood pressure and expression-derived signals.
- Browser tooling for evaluating the signal before integrating it.
Who it is for
- Product teams adding a physiological input to an existing application.
- Researchers who need a camera-based signal without instrumenting participants.
- Platforms exploring adaptive interfaces at scale.
Evidence
Built on the adaptive-interface patent family — US 10,423,893 (2019) and US 11,561,806 (2023).
Primary — Public patent records
Publicly documented API for camera-based rPPG signal estimation.
Company-reported — vitalsign.ai product documentation
What is not claimed
- Pulse-rate estimation works best under good lighting and limited motion. Higher-level signals are experimental and probabilistic.
- No accuracy percentage is published here, because no methodology, dataset or third-party evaluation is available to qualify one.
- Not a medical device. Health-adjacent use is a possible application only, and would require application-specific validation and appropriate regulatory review.
Work with VitalSign AI
Tell me what you are trying to build or measure. I read every message myself.