Form feedback

A method, comprising:

• detecting a spatio-temporal path of a handle during performance of an exercise movement involving use of a weight training machine;
• characterizing the spatio-temporal path of the handle including generating at least one parameter associated with the spatio-temporal path of the handle, wherein the spatio-temporal path of the handle is characterized at least in part by fusing together:
• (1) at least one of acceleration measurements or rate of rotation measurements from a first sensor comprising an inertial measurement unit embedded within the handle;
• and
• (2) cable length measurements from a second sensor different from the first sensor, wherein the cable length measurements from the second sensor are with respect to a cable coupled to the handle;
• detecting a sub-portion of a repetition of the exercise movement, wherein the sub-portion includes a first phase or a second phase;
• and
• comparing, based on the detected sub-portion, the at least one parameter associated with the detected spatio-temporal path of the handle to a canonical parameter, the canonical parameter comprising a threshold defined in a form detection expression corresponding to the exercise movement, wherein the at least one parameter includes a first parameter, a second parameter, or both, and wherein the comparing of the at least one parameter associated with the detected spatio-temporal path of the handle to the canonical parameter comprises:
• determining whether the detected sub-portion corresponds to the first phase or the second phase;
• in response to a determination that the detected sub-portion corresponds to the first phase, comparing the first parameter to the canonical parameter;
• and
• in response to a determination that the detected sub-portion corresponds to the second phase, comparing the second parameter to the canonical parameter.

A system, comprising:

• one or more processors configured to:
• determine a quality of a use of a weight training machine, wherein:
• the quality of the use of the weight training machine is determined based at least in part on a detected spatio-temporal path of a handle when performing an exercise movement;
• the quality of the use of the weight training machine is determined at least in part by evaluating a form detection expression corresponding to the exercise movement, wherein the form detection expression comprises at least one threshold;
• the evaluating of the form detection expression is based at least in part on fusing together:
• (1) at least one of acceleration measurements or rate of rotation measurements from a first sensor comprising an inertial measurement unit embedded within the handle;
• and
• (2) cable length measurements from a second sensor different from the first sensor, wherein the cable length measurements from the second sensor are with respect to a cable coupled to the handle;
• and
• the evaluating is performed with respect to a detected sub-portion of a repetition of the exercise movement, wherein the sub-portion includes a first phase or a second phase;
• the determining of the quality of the use of the weight training machine comprises to:
• compare, based on the detected sub-portion, the at least one parameter associated with the detected spatio-temporal path of the handle to a canonical parameter, the canonical parameter comprising a threshold defined in a form detection expression corresponding to the exercise movement, wherein the at least one parameter includes a first parameter, a second parameter, or both, and wherein the comparing of the at least one parameter associated with the detected spatio-temporal path of the handle to the canonical parameter comprises to:
• determine whether the detected sub-portion corresponds to the first phase or the second phase;
• in response to a determination that the detected sub-portion corresponds to the first phase, compare the first parameter to the canonical parameter;
• and
• in response to a determination that the detected sub-portion corresponds to the second phase, compare the second parameter to the canonical parameter;
• and
• perform an action that is based at least in part on the determined quality of the use of the weight training machine;
• and
• a memory coupled to the one or more processors and configured to provide the one or more processors with instructions.

A method, comprising:

• determining a quality of a use of a weight training machine, wherein:
• the quality of the use of the weight training machine is determined based at least in part on a detected spatio-temporal path of a handle when performing an exercise movement;
• the quality of the use of the weight training machine is determined at least in part by evaluating a form detection expression corresponding to the exercise movement, wherein the form detection expression comprises at least one threshold;
• the evaluating of the form detection expression is based at least in part on fusing together:
• (1) at least one of acceleration measurements or rate of rotation measurements from a first sensor comprising an inertial measurement unit embedded within the handle;
• and
• (2) cable length measurements from a second sensor different from the first sensor, wherein the cable length measurements from the second sensor are with respect to a cable coupled to the handle;
• and
• the evaluating is performed with respect to a detected sub-portion of a repetition of the exercise movement, wherein the sub-portion includes a first phase or a second phase;
• the determining of the quality of the use of the weight training machine comprises:
• comparing, based on the detected sub-portion, the at least one parameter associated with the detected spatio-temporal path of the handle to a canonical parameter, the canonical parameter comprising a threshold defined in a form detection expression corresponding to the exercise movement, wherein the at least one parameter includes a first parameter, a second parameter, or both, and
• the comparing of the at least one parameter associated with the detected spatio-temporal path of the handle to the canonical parameter comprises:
• determining whether the detected sub-portion corresponds to the first phase or the second phase;
• in response to a determination that the detected sub-portion corresponds to the first phase, comparing the first parameter to the canonical parameter;
• and
• in response to a determination that the detected sub-portion corresponds to the second phase, comparing the second parameter to the canonical parameter;
• and
• performing an action that is based at least in part on the determined quality of the use of the weight training machine.