1. An image processing apparatus, comprising:
an extractor to extract a body area from a color image based on a depth value of a depth image;
a matching unit to match a boundary of the extracted body area and a boundary of a generic body mesh model; and
a deformer to deform a mesh of the generic body mesh model based on a depth value of a pixel positioned within the boundary of the extracted body area.
2. The image processing apparatus of claim 1, wherein the extractor comprises:
a temporary extractor to extract a temporary body area from the color image based on the depth value of the depth image;
a classifier to classify the temporary body area into a foreground area, a background area, and an uncertain area based on the depth value; and
a determiner to determine and reclassify a pixel included in the uncertain area as a pixel included in one of the foreground area and the background area, based on a color value of the color image.
3. The image processing apparatus of claim 2, wherein the determiner performs the determination using a grab-cut algorithm.
4. The image processing apparatus of claim 1, wherein:
the extractor comprises:
a temporary extractor to extract a temporary body area from the color image based on the depth value of the depth image; and
a body area extractor to extract, as the body area, an area in which an area estimated as a head, an area estimated as a hand, and an area estimated as a foot are removed from the temporary body area, and
in the case of setting, as a neck, a point at which a horizontal length corresponding to a width of the temporary body area is absolutely short or is relatively shorter than an upper portion or a lower portion thereof,
the area estimated as the head is an area corresponding to an upper portion of the neck,
the area estimated as the hand corresponds to an end based on a first boundary between the hand and an arm that is obtained based on a color value difference of the color image, and
the area estimated as the foot corresponds to an end based on a second boundary between a leg and the foot that is obtained based on a color value difference of the color image.
5. The image processing apparatus of claim 4, wherein:
the extractor further comprises:
a feature point extractor to extract, as a feature point, at least one of both end points of the area estimated as the neck, both end points of the first boundary, both end points of the second boundary, and points of an armpit portion, and
the matching unit performs matching using the extracted feature point.
6. The image processing apparatus of claim 5, wherein the points of the armpit portion are detected based on a curvature calculation.
7. The image processing apparatus of claim 1, wherein the matching unit comprises:
a converter to generate a two-dimensional (2D) body mesh model by projecting the generic body mesh model; and
a body matching unit to match points of the 2D body mesh model to the boundary of the extracted body area while maintaining topology between the points of the 2D body mesh model.
8. The image processing apparatus of claim 7, wherein the body matching unit matches a feature point extracted from the boundary of the extracted body area and a feature point of the 2D body mesh model.
9. The image processing apparatus of claim 1, wherein the deformer matches a depth value within the boundary of the extracted body area and vertices of a 2D body mesh model, and deforms a mesh of the 2D body mesh model.
10. The image processing apparatus of claim 9, wherein the deformer deforms the mesh of the 2D body mesh model while maintaining topology between vertices of the 2D body mesh model.
11. An image processing method, comprising:
extracting a body area from a color image based on a depth value of a depth image;
matching a boundary of the extracted body area and a boundary of a generic body mesh model; and
deforming a mesh of the generic body mesh model based on a depth value of a pixel positioned within the boundary of the extracted body area.
12. The method of claim 11, wherein the extracting comprises:
extracting a temporary body area from the color image based on the depth value of the depth image;
classifying the temporary body area into a foreground area, a background area, and an uncertain area based on the depth value; and
determining and reclassifying a pixel included in the uncertain area as a pixel included in one of the foreground area and the background area, based on a color value of the color image.
13. The method of claim 11, wherein the extracting comprises:
extracting a temporary body area from the color image based on the depth value of the depth image;
in the case of setting, as a neck, a point at which a horizontal length corresponding to a width of the temporary body area is absolutely short or is relatively shorter than an upper portion or a lower portion thereof,
estimating, as a head, an area corresponding to an upper portion of the neck;
obtaining a first boundary between the hand and an arm based on a color value difference of the color image, and estimating, as a hand, an end based on the first boundary;
obtaining a second boundary between a leg and the foot based on a color value difference of the color image, and estimating, as a foot, an end based on the second boundary; and
extracting, as the body area, an area in which the area estimated as the head, the area estimated as the hand, and the area estimated as the foot are removed from the temporary body area.
14. The method of claim 13, wherein:
the extracting further comprises:
extracting, as a feature point, at least one of both end points of the area estimated as the neck, both end points of the first boundary, both end points of the second boundary, and points of an armpit portion, and
the matching comprises performing matching using the extracted feature point.
15. The method of claim 11, wherein the matching comprises:
generating a two-dimensional (2D) body mesh model by projecting the generic body mesh model; and
matching points of the 2D body mesh model to the boundary of the extracted body area while maintaining topology between the points of the 2D body mesh model.
16. The method of claim 15, wherein the matching of the points comprises matching a feature point extracted from the boundary of the extracted body area and a feature point of the 2D body mesh model.
17. The method of claim 11, wherein the deforming comprises matching a depth value within the boundary of the extracted body area and vertices of a 2D body mesh model, and deforming a mesh of the 2D body mesh model.
18. The method of claim 17, wherein the deforming comprises deforming the mesh of the 2D body mesh model while maintaining topology between vertices of the 2D body mesh model.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A game calling device, comprising:
a body having a mouthpiece end and a discharge end and multiple bores extending through the body from the mouthpiece end to the discharge end;
a first reed insert located in a first one of said bores;
a second reed insert located in a second one of said bores; and
a whistle insert located in a third one of said bores.
2. The game calling device according to claim 1, wherein said second reed insert is tuned for producing a higher pitch sound than said first reed insert.
3. The game calling device according to claim 2, wherein a mouthpiece having three separate inlet openings is located at said mouthpiece end of the body for selectively blowing air into the first, second and third bores, respectively.
4. The game calling device according to claim 3, wherein said inlet openings in the mouthpiece are spaced apart such that a user can selectively blow through each inlet individually or all three inlets simultaneously.
5. The game calling device according to claim 1, wherein said whistle is tuned to approximately 2500 Hz.
6. The game calling device according to claim 1, wherein said bores extend through the body in a diverging configuration relative to each other from the mouthpiece end toward the discharge end.
7. The game calling device according to claim 6, wherein said bores are generally coplanar with each other.
8. The game calling device according to claim 7, wherein said bores are angularly spaced about 15 degrees apart from each other.
9. The game calling device according to claim 1, wherein said body has a truncated fan-shape in plan view with lateral sides of the fan-shape emanating radially outwardly and diverging relative to each other from the mouthpiece end toward the discharge end, and the discharge end having a convex curvature providing a curved outer circumference of the fan-shape.
10. The game calling device according to claim 9, wherein a mouthpiece is located at the mouthpiece end of said body, said mouthpiece has three separate inlet openings for selectively blowing air into the first, second and third bores, respectively, and said mouthpiece has a concave curvature in plan view that corresponds to said convex curvature of the discharge end.
11. The game calling device according to claim 1, wherein said first and second reed inserts each have a tone board, a reed and a stopper assembled together to generate a desired sound, and said inserts further comprise O-rings fit into corresponding outer grooves for securing the inserts within respective bores of the body.
12. A game calling device, comprising:
a body having a mouthpiece end and a discharge end, and multiple bores extending through the body from the mouthpiece end to the discharge end, said body having a truncated fan-shape in plan view with lateral sides of the fan-shape emanating radially outwardly and diverging relative to each other from the mouthpiece end toward the discharge end; and
a plurality of vibratory inserts located in said bores for producing selected sounds as air is forced through the bores from the mouthpiece end to the discharge end.
13. The game calling device according to claim 12, wherein said multiple bores comprise first, second and third bores.
14. The game calling device according to claim 13, wherein a mouthpiece having first, second and third inlet openings is located at said mouthpiece end of the body for selectively blowing air into the first, second and third bores, respectively.
15. The game calling device according to claim 14, wherein said inlet openings in the mouthpiece are spaced apart such that a user can selectively blow through each inlet individually or all three inlets simultaneously.
16. The game calling device according to claim 15, wherein said first, second and third bores are generally coplanar with each other and diverge relative to each other from the mouthpiece end toward the discharge end.
17. The game calling device according to claim 16, wherein the discharge end of said body has a convex curvature in plan view that provides a curved outer circumference of the truncated fan-shape, and said mouthpiece has a concave curvature in plan view that provides a truncated apex portion of said truncated fan-shape.
18. The game calling device according to claim 12, wherein said vibratory inserts comprise a first reed insert for producing a first sound, a second reed insert for producing a second sound having a higher pitch than said first sound, and a whistle insert for producing a whistling sound having a higher pitch than said second sound.
19. A method of calling game animals, comprising the steps of:
providing a game calling device comprising:
a body having a mouthpiece end, a discharge end, and first, second and third bores extending through the body from the mouthpiece end to the discharge end,
first, second and third vibratory inserts placed in said first, second and third bores, respectively, for producing selected sounds as air is forced through the bores from the mouthpiece end to the discharge end, and
a mouthpiece having three separate inlet openings located at the mouthpiece end of the body for blowing air into the first, second and third bores, respectively; and
selectively blowing air into the three inlet openings to produce sounds having different pitches, with one of said sounds being a whistling sound.
20. The method of calling game animals according to claim 19, further comprising the steps of:
providing the body of the game calling device with a truncated fan-shape in plan view with lateral sides of the fan-shape emanating radially outwardly and diverging relative to each other from the mouthpiece end toward the discharge end; and
cupping a user’s hands over the lateral sides and discharge end of the body to control the sounds out of the game calling device.
21. The method of calling game animals according to claim 19, wherein the inlet openings in the mouthpiece are spaced apart from each other, and wherein said step of selectively blowing air into the three inlet openings comprises blowing air through said inlet openings in a selected sequence in which air is blown through a plurality of said inlet openings part of the time and through only one of the inlet openings part of the time.