Systems and Methods for Tracking a Model

• US 20120134575A1
• Filed: 02/02/2012
• Published: 05/31/2012
• Est. Priority Date: 05/29/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Mathe, Zsolt
• Granted Patent: US 8,351,652 B2
• Time in Patent Office: Days
• US Class Current: 382/154
• CPC Class Codes: G06T 17/05 Geographic models
• CPC Class Codes: G06V 40/23 Recognition of whole body movements, e.g. for sport training

Abstract

An image such as a depth image of a scene may be received, observed, or captured by a device and a model of a user in the depth image may be generated. The background of a received depth image may be removed to isolate a human target in the received depth image. A model may then be adjusted to fit with in the isolated human target in the received depth image. To adjust the model, a joint or a bone may be magnetized to the closest pixel of the isolated human target. The joint or the bone may then be refined such that the joint or the bone may be further adjusted to a pixels equidistant between two edges the body part of the isolated human target where the joint or bone may have been magnetized.

Claims

1. A method for tracking a model, the method comprising:

receiving a depth image of a scene;

removing a background of the depth image to isolate a target, wherein the target comprises a model having a plurality of parts; and

adjusting at least one of the parts of the model to fit within the isolated target.

2. The method of claim 1, wherein the model comprises a skeletal model having joints and bones.

3. The method of claim 2, wherein the adjusting at least one of the parts of the model to fit within the isolated target comprises adjusting a joint or a bone to fit within the isolated target.

4. The method of claim 2, wherein the adjusting at least one of the parts of the model comprises:

determining whether a joint of the model is outside an edge of the isolated target;

magnetizing the joint to a closest pixel; and

refining the joint within the isolated target based on the closest pixel.

5. The method of claim 4, wherein the magnetizing the joint to the closest pixel further comprises:

examining a set of pixels adjacent to a pixel in the depth image associated with a location of the joint to determine whether at least one of the pixels in the set of pixels has a valid depth value;

calculating distances between the pixel in the depth image associated with the location of the joint and each pixel in the set of pixels having valid depth values; and

selecting a shortest distance from the calculated distances as the closest pixel.

6. The method of claim 4, wherein the magnetizing the joint to the closest pixel further comprises assigning the joint an X value, a Y value, and a Z value associated with the closest pixel.

7. The method of claim 4, wherein the refining the joint within the isolated target based on the closest pixel further comprises determining a pixel equidistance from a first edge associated with the closest pixel and a second edge of the target associated with a pixel opposite of the closest pixel.

8. The method of claim 7, wherein the refining the joint within the isolated target based on the closest pixel further comprises assigning the portion of the model an X value, a Y, value, and a Z value of the pixel equidistance between the first edge of the target associated with the closest pixel and the second edge of the target associated with a pixel opposite of the closest pixel.

9. A computer-readable storage medium having stored thereon computer executable instructions for tracking a model, the computer executable instructions comprising instructions for:

receiving a depth image;

removing a background of the depth image to isolate a target;

determining whether a portion of the model is outside an edge of the isolated target;

and magnetizing the portion of the model outside the edge of the isolated target to a closest pixel of the isolated target.

10. The computer-readable storage medium of claim 9, wherein the model comprises a skeletal model having joints and bones.

11. The computer-readable storage medium of claim 10, wherein the portion of the model comprises at least one of the following:

a joint and a bone.

12. The computer-readable storage medium of claim 9, wherein the instructions for magnetizing the portion of the model outside the edge of the isolated target to the closest pixel of the isolated target further comprise instructions for:

examining a set of pixels adjacent to a pixel in the depth image associated with a location of the portion of the model outside the target to determine whether at least one of the pixels in the set of pixels has a valid depth value;

calculating distances between the pixel in the depth image associated with the location of the portion of the model and each pixel in the set of pixels having the valid depth value; and

selecting a shortest distance from the calculated distances as the closest pixel.

13. The computer-readable storage medium of claim 9, wherein the instructions for magnetizing the portion of the model outside the edge of the isolated target to the closest pixel of the isolated target further comprise instructions for assigning the portion of the model an X value, a Y value, and a Z value associated with the closest pixel.

14. The computer-readable storage medium of claim 9, further comprising instructions for refining the portion of the model within the isolated target.

15. The computer-readable storage medium of claim 14, wherein the instructions for refining the portion of the model within the isolated target further comprise instructions for assigning the portion of the model an X value, a Y, value, and a Z value of a pixel equidistance between a first edge of the target associated with the closest pixel and a second edge of the target associated with a pixel opposite of the closest pixel.

16. A system for tracking a model, the system comprising:

a capture device, wherein the capture device receives a depth image of a scene; and

a computing device in operative communication with the capture device, wherein the computing device comprises a processor that removes a background of the depth image to isolate a target;

determines whether a part of the model is outside an edge of the isolated target, and adjusts the part of the model outside the edge of the isolated target to a closest pixel of the isolated target.

17. The system of claim 16, wherein the processor adjusts the part of the model outside the edge of the isolated target to the closest pixel of the isolated target by assigning the part of the model an X value, a Y value, and a Z value associated with the closest pixel.

18. The system of claim 17, wherein the processor further refines the adjusted part of the model within the isolated target.

19. The system of claim 18, wherein the processor refines the adjusted part of the model within the isolated target by adjusting the part to be equidistant from a first edge associated with the closest pixel and a second edge of the target.