Creating a custom three-dimensional body shape model

1. A method, comprising:

receiving a series of captured images of a user;

extracting one or more silhouettes from the series of captured images of the user;

using a processor to process the captured images to identify a portion of each of the captured images corresponding to the user, wherein the identified portions of the captured images corresponding to the user include identified user joint locations of the user;

modifying, based on the identified user joint locations of the user, parameters of a predetermined three-dimensional human model to fit a modified version of the predetermined three-dimensional human model across the identified portions of the captured images to determine a set of specific parameters representing a body profile of the user; and

refining the modified version of the predetermined three-dimensional human model based on comparing the modified version of the predetermined three-dimensional human model to the one or more extracted silhouettes.

2. The method of claim 1 , wherein the parameters of the predetermined three-dimensional human model include one or more of the following: a height, a width, a leg length, and a hip width parameter.

3. The method of claim 1 , wherein the series of captured images of the user includes color data and depth data.

4. The method of claim 1 , wherein the identified portions of the captured images corresponding to the user include identified foreground portions of the captured images.

5. The method of claim 1 , wherein the user joint locations include shoulder, elbow, wrist, hand, hip, knee, ankle, and foot locations.

6. The method of claim 1 , wherein the identified portions of the captured images corresponding to the user include identified facial feature locations of the user.

7. The method of claim 6 , wherein the identified facial feature locations include ear locations, eye locations, a nose location, and a mouth location.

8. The method of claim 1 , further comprising determining a value identifying a predicted size fit between an item and the user based on the set of specific parameters representing the body profile of the user.

9. The method of claim 8 , further comprising selecting and providing the item to the user based on the value identifying the predicted size fit exceeding a threshold value.

10. The method of claim 1 , wherein the predetermined three-dimensional human model is based on a sex or an age of the user.

11. The method of claim 1 , wherein the predetermined three-dimensional human model includes a spectral decomposition model.

12. The method of claim 1 , further comprising:

identifying two-dimensional joint locations of the user in each of the captured images; and

fitting three-dimensional joint locations of the predetermined three-dimensional human model based on the identified two-dimensional joint locations.

13. The method of claim 1 , further comprising determining an accuracy metric of the modified version of the predetermined three-dimensional human model including by comparing rendered silhouettes of the modified version to silhouette images based on the identified portions of the captured images corresponding to the user.

14. The method of claim 1 , further comprising matching a stylist to the user based on attributes of the stylist matching the set of specific parameters representing the body profile of the user.

15. The method of claim 1 , further comprising:

selecting a subset of parameters of the set of specific parameters representing the body profile of the user; and

identifying a group of users with similar body profiles using the selected subset of parameters.

16. The method of claim 15 , further comprising at least in part automatically determining a parameter of a garment sized for the identified group of users.

17. The method of claim 1 , further comprising:

displaying a rendering of the modified version of the predetermined three-dimensional human model to the user; and

receiving feedback from the user regarding an accuracy rating of the modified version.

18. The method of claim 17 , wherein the feedback identifies a portion of the modified version where an additional refinement is desired.

19. The method of claim 18 , further comprising receiving one or more additional captured images of the user corresponding to the portion of the modified version where the additional refinement is desired.

20. A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving a series of captured images of a user;

extracting one or more silhouettes from the series of captured images of the user;

processing the captured images to identify a portion of each of the captured images corresponding to the user, wherein the identified portions of the captured images corresponding to the user include identified user joint locations of the user;

modifying, based on the identified user joint locations of the user, parameters of a predetermined three-dimensional human model to fit a modified version of the predetermined three-dimensional human model across the identified portions of the captured images to determine a set of specific parameters representing a body profile of the user; and

refining the modified version of the predetermined three-dimensional human model based on comparing the modified version of the predetermined three-dimensional human model to the one or more extracted silhouettes.

21. A system, comprising:

a processor; and

a memory coupled with the processor, wherein the memory is configured to provide the processor with instructions which when executed cause the processor to:

receive a series of captured images of a user;

extract one or more silhouettes from the series of captured images of the user;

process the captured images to identify a portion of each of the captured images corresponding to the user, wherein the identified portions of the captured images corresponding to the user include identified user joint locations of the user; and

modify, based on the identified user joint locations of the user, parameters of a predetermined three-dimensional human model to fit a modified version of the predetermined three-dimensional human model across the identified portions of the captured images to determine a set of specific parameters representing a body profile of the user; and

refine the modified version of the predetermined three-dimensional human model based on comparing the modified version of the predetermined three-dimensional human model to the one or more extracted silhouettes.