System and method for improving exercise performance using a mobile device

A computer system comprising one or more processors, one or more memories, and one or more computer-readable hardware storage devices, the one or more computer-readable hardware storage devices containing program code executable by the one or more processors via the one or more memories to implement a sensor-less method for improving exercise performance, the method comprising:

• receiving video exercise data captured by a camera of the computing device, the video exercise data being associated with a user performing an active exercise and wearing no sensors;
• converting the video exercise data via a deep learning algorithm to one or more movements of a user's skeleton performing the active exercise in real-time;
• extracting parameters associated with a first movement of the one or more movements from an external database in a post-processing stage, wherein the parameters associated with the first movement comprise at least an angle of a bone of the user engaging in the first movement in relation to the camera and an orientation of the bone engaging in the first movement in relation to the camera;
• comparing the first movement of the one or more movements to a predictive model associated with an ideal performance of the first movement;
• assigning a confidence score to the comparison;
• and
• transmitting feedback data to a graphical user interface of the computing device for display to the user.

A sensor-less method executed by a motion analysis engine of a computing device for improving exercise performance, the sensor-less method comprising:

• receiving video exercise data captured by a camera of the computing device, the video exercise data being associated with a user performing an active exercise and wearing no sensors;
• converting the video exercise data via a deep learning algorithm to one or more movements of a user's skeleton performing the active exercise in real-time;
• extracting parameters associated with a first movement from an external database in a post-processing stage, wherein the parameters associated with the first movement comprise at least an angle of a bone of the user engaging in the first movement in relation to the camera and an orientation of the bone engaging in the first movement in relation to the camera;
• comparing the first movement of the one or more movements to a predictive model associated with an ideal performance of the first movement;
• assigning a confidence score to the comparison;
• and
• transmitting feedback data to a graphical user interface of the computing device for display to the user.