Dominant color extraction for ambient light derived from video content mapped through unrendered color space
View Patent ↗Extracting and processing video content encoded in a rendered color space to be emulated by an ambient light source, comprising extracting dominant color information from a video signal and transforming the color information through unrendered color space using tristimulus primary matrices to form a second rendered color space for ambient distribution. Steps include quantizing the rendered color space to form an assigned color distribution, such as by reducing possible color states, or binning to form superpixels; and selecting a dominant color using a mode, mean, median, or weighted average of pixel chromaticities. A color of interest can be further analyzed to produce a true dominant color, and past video frames can guide selection of dominant colors in future frames.
1. A method for dominant color extraction from video content encoded in a rendered color space (RGB) to produce a dominant color (DC) for emulation by an ambient light source ( 88 ), the method comprising the steps of:
(1) quantizing at least some pixel chromaticities (Cp) from said video content in said rendered color space to form a distribution of assigned colors (AC);
(2) performing dominant color extraction from said distribution of assigned colors to produce a dominant color by extracting any of: (a) a mode of said assigned colors, (b) a median of said assigned colors, (c) a weighted average by chromaticity of said assigned colors, and (d) a weighted average using a weighting function (W(i,j,R));
(3) transforming said dominant color from said rendered color space to a second rendered color space (R′G′B′) so formed as to allow driving said ambient light source.
2. The method as claimed in claim 1 , wherein said quantizing comprises assigning a plurality of said pixel chromaticities (Cp) to one of said assigned colors.
3. The method as claimed in claim 1 , wherein said quantizing comprises binning said pixel chromaticities into at least one superpixel (XP).
4. The method as claimed in claim 3 , wherein any of a size, orientation, shape or location of said superpixel is formed in conformity with an image feature (J 8 ).
5. The method as claimed in claim 3 , wherein said method additionally comprises the step of:
assigning a plurality of said pixel chromaticities (Cp) in said superpixel to one of said assigned colors.
6. The method as claimed in claim 1 , wherein at least one of said assigned colors is a regional color vector (V) that is not necessarily in said rendered color space.
7. The method as claimed in claim 6 , wherein said regional color vector lies in said second rendered color space.
8. The method as claimed in claim 1 , wherein said method additionally comprises the step of:
establishing at least one color of interest (COI) in said distribution of assigned colors and extracting pixel chromaticities assigned thereto to derive a true dominant color (TDC) to be designated as said dominant color.
9. The method as claimed in claim 8 , wherein said pixel chromaticities are, at least in part, derived from a superpixel (XP) obtained by binning said pixel chromaticities.
10. The method as claimed in claim 1 , wherein said dominant color comprises a pallet of dominant colors (DC 1 +DC 2 +DC 3 ).
11. The method as claimed in claim 1 , wherein said transforming step (3) comprises:
(3a) transforming said dominant color from said rendered color space to an unrendered color space (XYZ); and
(3b) transforming said dominant color from said unrendered color space to said second rendered color space.
12. The method as claimed in claim 11 , wherein said step (3) of transforming additionally comprises:
(3c) matrix transformations of primaries (RGB, R′G′B′) of said rendered color space and second rendered color space to said unrendered color space using first and second tristimulus primary matrices (M 1 , M 2 ); and
deriving a transformation of said color information into said second rendered color space (R′G′B′) by matrix multiplication of said primaries of said rendered color space, said first tristimulus matrix, and the inverse of said second tristimulus matrix (M 2 ) −1 .
13. The method as claimed in claim 1 , wherein the pixel chromaticities of the quantizing step (1) are obtained from an extraction region (R 1 ), and wherein said method additionally comprises the step of:
(4) broadcasting ambient light (L 1 ) of said dominant color from said ambient light source adjacent said extraction region.
14. The method as claimed in claim 1 , wherein the quantizing step (1) additionally comprises decoding said video signal into a set of frames, and wherein a plurality of pixel chromaticities are obtained from an image feature (J 8 ) extracted (FE) from one frame of said set of frames.
15. The method as claimed in claim 1 , wherein said quantizing step (1) additionally comprises decoding said video signal into a set of frames, and wherein the dominant color of a frame (F 4 ) is obtained at least in part by relying upon at least one dominant color from a previous frame (F 3 ).
16. A method for dominant color extraction from video content encoded in a rendered color space (RGB) to produce a dominant color (DC) for emulation by an ambient light source ( 88 ), the method comprising the steps of:
(1) decoding said video content in said rendered color space into a plurality of frames, and quantizing at least some pixel chromaticities (Cp) from an extraction region (R 1 ) in one of said frames to form a distribution of assigned colors (AC);
(2) performing dominant color extraction from said distribution of assigned colors to produce a dominant color by extracting any of: (a) a mode of said assigned colors, (b) a median of said assigned colors, (c) a weighted average by chromaticity of said assigned colors, and (d) a weighted average using a weighting function (W(i,j,R));
(3a) transforming said dominant color from said rendered color space to an unrendered color space (XYZ);
(3b) transforming said dominant color from said unrendered color space to said second rendered, color space, assisted by
(3c) matrix transformations of primaries (RGB, R′G′B′) of said rendered color space and second rendered color space to said unrendered color space using first and second tristimulus primary matrices (M 1 , M 2 ); and
deriving a transformation of said color information into said second rendered color space (R′G′B′) by matrix multiplication of said primaries of said rendered color space, said first tristimulus matrix, and the inverse of said second tristimulus matrix (M 2 ) −1 ; and
(4) broadcasting ambient light (L 1 ) of said dominant color from said ambient light source adjacent said extraction region.
17. The method as claimed in claim 16 , wherein said extraction region is chosen to be an image feature (J 8 ) extracted (FE) from one frame.
18. The method as claimed in claim 16 , wherein said method additionally comprises the step of:
establishing at least one color of interest (COI) in said distribution of assigned colors and extracting pixel chromaticities assigned thereto to derive a true dominant color (TDC) to be designated as said dominant color.
19. The method as claimed in claim 16 , wherein at least one of said assigned colors is a regional color vector (V) that is not necessarily in said rendered color space.
20. The method as claimed in claim 16 , wherein said frames include a first and a second frame, and the dominant color of a frame (F 4 ) is obtained at least in part by relying upon at least one dominant color from a previous frame (F 3 ).