Initializing orientation in space for predictive information for free space gesture control and communication
View Patent ↗The technology disclosed relates to initializing orientation of a three-dimensional (3D) model of an object. In particular, it relates to accessing at least one three-dimensional (3D) model of an object and observed information of the object movable in space and determining a primary orientation parameter of the model from the observed information. The method further includes detecting contours of the object in the observed information and calculating a representative normal to the detected contours, accessing a vector representing a 3D angle from the object to a point of observation, calculating a primary orientation of the object as a cross-product of the representative normal and the vector, and using the calculated primary orientation parameter to initialize the model.
1. A computer implemented method of initializing an orientation of a three-dimensional (3D) model of an object, comprising:
accessing at least one three-dimensional (3D) model of the object and observed information of the object moving in space; and
determining a primary orientation parameter for the 3D model from the observed information, including:
detecting a plurality of contours for the object in the observed information and calculating a representative normal vector for at least some of the detected contours;
accessing a tangent vector representing a 3D angle tangent to a point on a surface of the object and drawn to a point of observation;
calculating a primary orientation parameter for the object using a cross-product of the representative normal vector and the tangent vector; and
using the calculated primary orientation parameter to initialize the orientation of the 3D model of the object.
2. The method of claim 1 , further including filtering the plurality of contours to detect edges and calculating the representative normal vector from the filtered contours.
3. The method of claim 1 , further including using brightness gradients to calculate the representative normal vector.
4. The method of claim 1 , further including using brightness contours to calculate the representative normal vector.
5. The method of claim 1 , wherein the calculating the representative normal vector includes averaging a plurality of normal vectors.
6. The method of claim 5 , further including preparing a histogram of the plurality of normal vectors; calculating an argument value for the histogram; and using the argument value calculating the representative normal vector.
7. The method of claim 1 , further including:
identifying a plurality of normal vectors normal to at least some of the contours of the object and determining for a selected contour, a representative normal vector from the plurality of normal vectors associated with the selected contour based upon a magnitude computed for the plurality of normal vectors; and
selecting from among the plurality of normal vectors for the selected contour a normal vector having a maximum magnitude as the representative normal vector for the selected contour.
8. The method of claim 1 , further including:
identifying a plurality of normal vectors normal to at least some of the contours of the object; and
selecting, as the representative normal vector, a normal vector selected from the plurality of normal vectors, the selected representative normal vector having a magnitude determined from magnitudes of the plurality of normal vectors and a direction determined from the plurality of normal vectors.
9. The method of claim 8 , the determining further includes determining an average of the magnitudes and the directions of the plurality of normal vectors.
10. The method of claim 8 , the determining further includes determining a mean of the magnitudes and the directions of the plurality of normal vectors.
11. The method of claim 1 , further including initializing a model portion by applying the primary orientation parameter to at least a portion of the model and aligning the portion to an observed orientation of the object.
12. The method of claim 1 , wherein the object is a human body.
13. The method of claim 1 , wherein the object is a hand.
14. The method of claim 1 , wherein the object is a tool.
15. The method of claim 1 , further including modifying power consumption profiles of a machine sensory control system responsive to initialization of the model.
16. A system of initializing an orientation of a three-dimensional (3D) model of an object, comprising:
a processor and a computer readable storage medium storing computer instructions configured to cause the processor to:
access at least one three-dimensional (3D) model of the object and observed information of the object moving in space; and
determine a primary orientation parameter for the 3D model from the observed information, including:
detecting a plurality of contours for the object in the observed information and calculating a representative normal vector for at least some of the detected contours;
accessing a tangent vector representing a 3D angle tangent to a point on a surface of the object and drawn to a point of observation;
calculating a primary orientation parameter for the object using a cross-product of the representative normal vector and the tangent vector; and
using the calculated primary orientation parameter to initialize the orientation of the 3D model of the object.
17. The system of claim 16 , further configured to filter the plurality of contours to detect edges and calculating the representative normal vector from the filtered contours.
18. The system of claim 16 , further configured to use brightness gradients to calculate the representative normal vector.
19. The system of claim 16 , further configured to use brightness contours to calculate the representative normal vector.
20. The system of claim 16 , further configured to average a plurality of normal vectors.
21. The system of claim 20 , further configured to prepare a histogram of the plurality of normal vectors; calculate an argument value for the histogram; and using the argument value as the representative normal vector.
22. A non-transitory computer readable storage medium impressed with computer program instructions to initialize an orientation of a three-dimensional (3D) model of an object, the instructions, when executed on a processor, implement a method comprising:
accessing at least one three-dimensional (3D) model of the object and observed information of the object moving in space; and
determining a primary orientation parameter for the 3D model from the observed information, including:
detecting a plurality of contours for the object in the observed information and calculating a representative normal vector for at least some of the detected contours;
accessing a tangent vector representing a 3D angle tangent to a point on a surface of the object and drawn to a point of observation;
calculating a primary orientation parameter for the object using a cross-product of the representative normal vector and the tangent vector; and
using the calculated primary orientation parameter to initialize the orientation of the 3D model of the object.