Re-coloring a color image

1. A method for re-coloring a region of a color image comprising:

determining an original base color in the region of the color image by implementing an ambient/diffuse shading model using a processing unit

establishing at least one shading parameter in the region of the color image by implementing the ambient/diffuse shading model using the processing unit;

combining a new base color with the at least one shading parameter in the region of the color image thereby producing a re-colored region of the color image using the processing unit,

wherein the ambient/diffuse shading model is implemented according to:

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, or

E total =E light E base ( k ambient +k diffuse cos θ)

which under the assumptions

E light =1

k ambient +k diffuse =1

simplifies to

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, or

E i =E base (1+ k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, and where i represents a pixel within the region, θ i being the angle formed at pixel i, or

E c,i =E c,base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E c,i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, where i represents a pixel within the region, θ i being the angle formed at pixel i, and where the subscript c indicates a color channel, or

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and,

wherein the base color E base is the original base color and is determined by identifying the maximum color of the region, where the surface normal is pointing directly toward the light source so that θ=0 and cos θ=1, and:

E

base

=

E

maximum

1

+

k

diffuse

⁡

(

cos

⁡

(

0

)

-

1

)

=

E

maximum

,

or

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and,

wherein the diffuse coefficient k diffuse is determined by identifying the minimum color of the region, where the surface normal is pointing away from the light source so that θ=π/2 and cos θ=0, and:

E minimum =E base (1 +k diffuse (cos(π/2)−1))

from which k diffuse is solved:

k diffuse =( E base −E minimum )/ E base , or

E i =E base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E i is the total color, E base represents the base color, the term k diffuse (cos θ i −1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, and where i represents a pixel within the region, θ i being the angle formed at pixel i; and,

wherein for each pixel i in the region, the value for cos θ i is computed from the shading model:

E i =E base (1 +k diffuse (cos θ i −1))

which may be solved for cos θ i and simplified to:

cos θ i =( E i −E minimum )/( E base −E minimum ), or

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and,

wherein a new base color, E base ′ is substituted into the shading equation for each pixel in the region, producing a re-colored second region:

E total ′=E base ′(1 +k diffuse (cos θ−1))

where the prime, ′, refers to the new base color, as opposed to the original base color.

2. A method for re-coloring according to claim 1 wherein the at least one shading parameter is established separate from the original base color.

3. A method according to claim 1 wherein when the ambient/diffuse shading model is implemented according to E total =E base (1+k diffuse (cos θ−1)) the energy value, E base , is computed according to one or more of red, green and blue values.

4. A method for re-coloring according to claim 1 wherein the original and new base colors include color components and the color components are transformed into and out of a three-dimensional, linear energy color space.

5. A method for re-coloring according to claim 1 wherein the original and new base colors include color components and the color components are transformed into and out of a three-dimensional, linear energy color space, the linear energy space including at least one of linear energy RGB and CIE XYZ.

6. A method for re-coloring according to claim 1 wherein the new base color is obtained from at least one of a two-dimensional texture map and a procedural source.

7. A method according to claim 1 wherein the original base color includes color components and the at least one shading parameter is computed from one or more of the color components.

8. A method according to claim 1 wherein the original base color includes color components and the at least one shading parameter is computed from the color component showing the largest variation.

9. A computer program product in a non-transitory computer-readable medium for use in a data processing system for executing a computer program, the computer program product comprising instructions for operating a method according to claim 1 .

10. A computer program product in a non-transitory computer-readable medium for use in a data processing system for executing a computer program, the computer program product comprising:

first instructions for determining an original base color in a region of a color image based on an ambient/diffuse shading model;

second instructions for establishing at least one shading parameter in the region of the color image based on the ambient/diffuse shading model; and

third instructions for combining a new base color with the at least one shading parameter in the region of the color image thereby producing a re-colored region of the color image,

wherein the ambient/diffuse shading model is implemented according to:

E total =E base (1+ k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, or

E total =E light E base ( k ambient +k diffuse cos θ)

which under the assumptions E light =1

k ambient +k diffuse =1

simplifies to

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, or

E i =E base (1+ k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, and where i represents a pixel within the region, θ i being the angle formed at pixel i, or

E c,i =E c,base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E c,i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, and where i represents a pixel within the region, θ i being the angle formed at pixel I, and where the subscript c indicates a color channel, or

E total =E base (1+ k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and

wherein the base color E base is the original base color and is determined by identifying the maximum color of the region, where the surface normal is pointing directly toward the light source so that θ=0 and cos θ=1, and:

E

base

=

E

maximum

/

1

+

k

diffuse

⁡

(

cos

⁡

(

0

)

-

1

)

=

E

maximum

,

or

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and

wherein the diffuse coefficient k diffuse is determined by identifying the minimum color of the region, where the surface normal is pointing away from the light source so that θ=π/2 and cos θ=0, and:

E minimum =E base (1 +k diffuse (cos(π/2)−1))

from which k diffuse is solved:

k diffuse =( E base −E minimum )/ E base , or

E i =E base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, and where i represents a pixel within the region, θ i being the angle formed at pixel i, and,

wherein for each pixel i in the region, the value for cos θ i is computed from the shading model:

E i =E base (1 +k diffuse (cos θ i −1))

which may be solved for cos θ i and simplified to:

cos θ i =( E i −E minimum )/( E base −E minimum ), or

E total =E base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and,

wherein a new base color E′ base , is substituted into the shading equation for each pixel in the region, producing a re-colored second region:

E′ total =E′ base (1+ k diffuse (cos θ i −1))

where the prime, ′, refers to the new base color, as opposed to the original base color.

11. A method for identifying an original base color of a region of a color image, the method comprising:

evaluating the color content of the region of the color image by implementing an ambient/diffuse shading model using a processing unit; and,

separating a set of one or more shading parameters for the region of the color image from the original base color by implementing the ambient/diffuse shading model using the processing unit;

wherein the ambient/diffuse shading model is implemented according to:

E total =E base (1 +k diffuse (cos θ−1)),

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, or

E total =E light E base ( k ambient +k diffuse cos θ)

which under the assumptions E light =1

k ambient +k diffuse =1

simplifies to

E total =E base (1 +k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, or

E i =E base (1+ k diffuse (cos θ−1))

where E is the energy relationship of light and color in the energy domain, E i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, and where i represents a pixel within the region, θ i being the angle formed at pixel i, or

E c,i =E c,base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E c,i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, where i represents a pixel within the region, θ i being the angle formed at pixel i, and where the subscript c indicates a color channel, or

E total =E base (1 +k diffuse (cos θ−1)),

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and,

wherein the base color E base is the original base color and is determined by identifying the maximum color of the region, where the surface normal is pointing directly toward the light source so that θ=0 and cos θ=1, and:

E

base

=

E

maximum

1

+

k

diffuse

⁡

(

cos

⁡

(

0

)

-

1

)

=

E

maximum

,

or

E total =E base (1 +k diffuse (cos θ−1)),

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and,

wherein the diffuse coefficient k diffuse is determined by identifying the minimum color of the region, where the surface normal is pointing away from the light source so that θ=π/2 and cos θ=0, and:

E minimum =E base (1 +k diffuse (cos(π/2)−1))

from which k diffuse is solved:

k diffuse =( E base −E minimum )/ E base , or

E i =E base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E i is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image, and where i represents a pixel within the region, θ i being the angle formed at pixel i; and,

wherein for each pixel i in the region, the value for cos θ i is computed from the shading model:

E i =E base (1 +k diffuse (cos θ i −1))

which may be solved for cos θ i and simplified to:

cos θ i =( E i −E minimum )/( E base −E minimum ), or

E total =E base (1 +k diffuse (cos θ i −1))

where E is the energy relationship of light and color in the energy domain, E total is the total color, E base represents the base color, the term k diffuse (cos θ−1) represents the at least one shading parameter and θ is the angle formed between a surface normal in the region and a corresponding vector pointing toward a light source in the image; and,

wherein a new base color, E base ′ is substituted into the shading equation for each pixel in the region, producing a re-colored second region:

E total ′=E base ′(1+ k diffuse (cos θ−1))

where the prime, ′, refers to the new base color, as opposed to the original base color.

12. A method according to claim 11 wherein the separating operation is a mathematical representation of the base color and the one or more shading parameters.