Scoring metric for physical activity performance and tracking
Certain aspects of the present disclosure provide a method for assessing the performance of a physical activity, including: recording motion capture data while a training subject demonstrates a physical activity sequence; identifying one or more primary body elements based on the motion capture data; deriving one or more path characteristic metrics based on a state variable set and the one or more primary body elements, wherein the state variable set defines the state of a body at any given time; and defining an ideal activity path of the physical activity sequence based on the one or more path characteristic metrics.
1. An apparatus comprising:
at least one processor;
a memory coupled to the at least one processor, the memory comprising instructions that, when executed by the at least one processor, cause the at least one processor to:
access training motion capture data of a plurality of samples of a body of at least one training subject performing a physical activity sequence comprising a sequence of states,
determine at least one primary body element for the physical activity sequence, and
determine a path tracking model configured to numerically represent a path of the at least one primary body element as the body performs the physical activity sequence, the path tracking model determined via:
determining at least one path characteristic metric of the at least one primary body element during each of the sequence of states, the at least one path characteristic metric comprising a numeric variable defining state information of the at least one primary body element during one of the sequence of states, and
determining an ideal activity path of the physical activity sequence, the ideal activity path comprising an ordered sequence of the at least one path characteristic metric during the sequence of states normalized with respect to time for each of the plurality of samples to associate the ideal activity path with a time component expressed as a percentage of completion.
2. The apparatus of claim 1 , the state information of the at least one primary body element comprising at least one of a position or an orientation of the at least one primary body element.
3. The apparatus of claim 1 , the instructions, when executed by the at least one processor, to cause the at least one processor to:
quantify work performed by a plurality of body elements in the training motion capture data, and
select at least one of the plurality of body elements performing work above a threshold as the at least one primary body element.
4. The apparatus of claim 1 , the instructions, when executed by the at least one processor, to cause the at least one processor to determine a plurality of key states of the physical activity sequence based on the training motion capture data, each of the plurality of key states comprising a specific state that in-part defines the physical activity sequence.
5. The apparatus of claim 4 , the ideal activity path comprising a sequence of the plurality of key states.
6. The apparatus of claim 1 , the path tracking model comprising at least one path similarity measure numerically quantifying a similarity between an actual activity path and the ideal activity path.
7. The apparatus of claim 1 , the at least one path characteristic metric comprising a unit vector of the primary body element.
8. The apparatus of claim 1 , the at least one path characteristic metric associated with temporal information indicating an amount of time to complete at least a portion of the ideal activity path.
9. The apparatus of claim 1 , the ideal activity path comprising a multi-dimensional path of the at least one primary body element between each of the sequence of states, each dimension of the multi-dimensional path comprising a spatial dimension associated with the at least one primary body element.
10. The apparatus of claim 1 , the ideal activity path further defined based on multiple recorded physical activity sequence samples of different performances of the physical activity sequence captured in the training motion capture data.
11. A computer-implemented method, comprising, via at least one processor of a computing device:
accessing training motion capture data of a plurality of samples of a body of at least one training subject performing a physical activity sequence comprising a sequence of states,
determining at least one primary body element for the physical activity sequence, and
determining a path tracking model configured to numerically represent a path of the at least one primary body element as the body performs the physical activity sequence, the path tracking model determined via:
determining at least one path characteristic metric of the at least one primary body element during each of the sequence of states, the at least one path characteristic metric comprising a numeric variable defining state information of the at least one primary body element during one of the sequence of states, and
determining an ideal activity path of the physical activity sequence, the ideal activity path comprising an ordered sequence of the at least one path characteristic metric during the sequence of states normalized with respect to time for each of the plurality of samples to associate the ideal activity path with a time component expressed as a percentage of completion.
12. The method of claim 11 , the state information of the at least one primary body element comprising at least one of a position or an orientation of the at least one primary body element.
13. The method of claim 11 , comprising:
quantifying work performed by a plurality of body elements in the training motion capture data, and
selecting at least one of the plurality of body elements performing work above a threshold as the at least one primary body element.
14. The method of claim 11 , comprising determining a plurality of key states of the physical activity sequence based on the training motion capture data, each of the plurality of key states comprising a specific state that in-part defines the physical activity sequence.
15. The method of claim 14 , the ideal activity path comprising a sequence of the plurality of key states.
16. The method of claim 11 , the path tracking model comprising at least one path similarity measure numerically quantifying a similarity between an actual activity path and the ideal activity path.
17. The method of claim 11 , the at least one path characteristic metric comprising a unit vector of the primary body element.
18. The method of claim 11 , the at least one path characteristic metric associated with temporal information indicating an amount of time to complete at least a portion of the ideal activity path.
19. The method of claim 11 , the ideal activity path comprising a multi-dimensional path of the at least one primary body element between each of the sequence of states, each dimension of the multi-dimensional path comprising a spatial dimension associated with the at least one primary body element.
20. The method of claim 11 , the ideal activity path further defined based on multiple recorded physical activity sequence samples of different performances of the physical activity sequence captured in the training motion capture data.