/AI-Fitness-SDK

Integrate your personal AI Fitness Trainer in minutes with our SDK

Primary LanguageJavaScriptApache License 2.0Apache-2.0

INTEGRATE AI TRAINER IN YOUR APP IN MINUTES

Easily transform your platform with our SDK: white-labeled workouts with precise motion tracking and real-time feedback tailored for accuracy and engagement

Jump to an example in your programming language:

Flutter React Native Swift Kotlin HTML & JS React Progressive Web App
0b36f063-376b-4c30-a543-9112c4dc7e92.mp4

KinesteX revolutionizes fitness and health industries with our cutting-edge embeddable SDK. Using device's camera, our AI precisely analyzes human movements in real-time, providing instant, expert-level guidance tailored to the individual's form. Our movements recognition models consistently exceed 90% accuracy, enhancing user's workout experience, and helping them achieve their fitness goals faster. Our AI Fitness Trainer is available for your programming language, we have created light-weight packages for Flutter, SwiftUI, React Native, Kotlin, PWA. In case your programming language is missing, please contact us and we will create a custom example for you and guide you through the process!

Available data types

Type Data Description
kinestex_launched Format: dd mm yyyy hours:minutes:seconds When a user has launched KinesteX
exit_kinestex Format: date: dd mm yyyy hours:minutes:seconds, time_spent: number Logs when a user clicks on exit button, requesting dismissal of KinesteX and sending how much time a user has spent totally in seconds since launch
plan_unlocked Format: title: String, date: date and time Logs when a workout plan is unlocked by a user
workout_opened Format: title: String, date: date and time Logs when a workout is opened by a user
workout_started Format: title: String, date: date and time Logs when a workout is started.
error_occurred Format: data: string Logs when a significant error has occurred. For example, a user has not granted access to the camera
exercise_completed Format: time_spent: number, repeats: number, calories: number, exercise: string, mistakes: [string: number] Logs everytime a user finishes an exercise
total_active_seconds Format: number Logs every 5 seconds and counts the number of active seconds a user has spent working out. This value is not sent when a user leaves camera tracking area
left_camera_frame Format: number Indicates that a user has left the camera frame. The data sent is the current number of total_active_seconds
returned_camera_frame Format: number Indicates that a user has returned to the camera frame. The data sent is the current number of total_active_seconds
workout_overview Format: workout: string,total_time_spent: number, total_repeats: number, total_calories: number, percentage_completed: number, total_mistakes: number Logged when a user finishes the workout with a complete short summary of the workout
exercise_overview Format: [exercise_completed] Returns a log of all exercises and their data (exercise_completed data is defined 5 lines above)
workout_completed Format: workout: string, date: dd mm yyyy hours:minutes:seconds Logs when a user finishes the workout and exits the workout overview
active_days (Coming soon) Format: number Represents a number of days a user has been opening KinesteX
total_workouts (Coming soon) Format: number Represents a number of workouts a user has done since start of using KinesteX
workout_efficiency (Coming soon) Format: number Represents the level of intensivity a person has done the workout with. An average level of workout efficiency is 0.5, which represents an average time a person should complete the workout for at least 80% within a specific timeframe. For example, if on average people complete workout X in 15 minutes, but a person Y has completed the workout in 12 minutes, they will have a higher workout_efficiency number

To get demo access and your api key, please fill out a form on our website

Or, alternatively:

  • Send us an email to support@kinestex.com
  • In the email subject write: B2B Demo Access
  • In the email body include any supporting details (company / project name, site, programming language your project is written in, etc.)

Once you have contacted us, you may

Refer to packages based on your programming language:

Flutter React Native Swift Kotlin HTML & JS React Progressive Web App

For any questions, please contact: support@kinestex.com