One-page planner: the three AI feature groups (personalize the plan / watch the movement / read the body), the two lines (estimation: a camera estimates, never measures; wellness: claims set medical-device status), where each model runs (phone vs cloud vs wearable layer), the 33 ms real-time frame budget, the camera-placement rule (0% to 95.5% squat detection by position, per the 2026 JMIR study), the embed-vs-assemble-vs-build split, and a privacy & compliance gate (general-wellness claims discipline, GDPR special-category, FTC Health Breach rule, platform health policies).
Download free PDF
The handbook · 7 volumes
Fora Soft Video Engineering Handbook
This planner sorts fitness app AI features into three groups, sets where each model runs within a 33 ms frame budget, and adds a wellness and privacy compliance gate. Our seven-volume handbook covers this stack; Volume 7 covers real-time AI pipelines, serving and regulation, Volume 3 covers WebRTC and players, and Volume 1 covers encoding. Written by Nikolay Sapunov, CEO at Fora Soft.
Adding pose tracking or body metrics to a fitness app without medical-device risk?
Describe your AI features, and we will decide what runs on phone or cloud and keep your claims inside wellness rules. 250+ video products shipped since 2005.
You get a realistic estimate, technical recommendations and next steps. Free, no obligations.
