YCgCo

The YCgCo color model (or YCoCg) describes the color space formed by luminance Y, chrominance green Cg, and chrominance orange Co. This is supported by video compression codecs such as H.264/MPEG-4 AVC and Dirac^[1] since it can losslessly be converted to and from RGB with fewer bits required compared with other color models.

Original image above and representation of the individual components Y, chrominance green Cg and chrominance orange Co.

Comparison with other color models

RGB color model

The three values of the YCgCo color model are calculated as follows from the three color values of the RGB color model:

$\begin{bmatrix} Y \\ Cg \\ Co \end{bmatrix} = \begin{bmatrix} 1/4 & 1/2 & 1/4\\ -1/4 & 1/2 & -1/4\\ 1/2 & 0 & -1/2\end{bmatrix} \cdot \begin{bmatrix} R \\ G \\ B \end{bmatrix}$

The values of the luminance Y are in the range from 0 to 1, while chrominance green Cg and chrominance orange Co are in the range of −0.5 to 0.5. For example, pure red is expressed in the RGB system as (1, 0, 0) and in the YCgCo system as (1/4, −1/4, 1/2).^[2][3]

The inverse matrix converts from the YCgCo color model back to the RGB color model:

$\begin{bmatrix} R \\ G \\ B \end{bmatrix} = \begin{bmatrix} 1 & -1 & 1\\ 1 & 1 & 0\\ 1 & -1 & -1\end{bmatrix} \cdot \begin{bmatrix} Y \\ Cg \\ Co \end{bmatrix}$

To perform the conversion, only two additions and two subtractions are necessary, and there are no real coefficients. The transformation can be efficiently implemented as integer additions and subtractions:

tmp := Y   − Cg;
R   := tmp + Co;
G   := Y   + Cg;
B   := tmp − Co;

YCbCr color model

An advantage the YCgCo color model has over the YCbCr color model is simpler and faster computation, and better decorrelation of the color planes to improve compression performance.^[2][3]

Literature

Textbook:

• Tilo Strutz, Image data compression, Foundations, coding, wavelets, JPEG, MPEG, H.264 4th edition, Vieweg+Teubner 2009, ISBN 978-3-8348-0472-3 (Print), ISBN 978-3-8348-9986-6 (Online)

Research on YCgCo color model:

• Henrique Malvar and Gary Sullivan, "YCoCg-R: A color space with RGB reversibility and low dynamic range". Moving Picture Experts Group and Video Coding Experts Group, document JVT-I014, JVT PExt Ad Hoc Group Meeting: Trondheim, Norway, July 2003.
• Shijun Sun, "Residual Color Transform Using YCoCg-R". ISO/IEC JTC1/SC29/WG11 and ITU-T Q6/SG16, document JVT-L014, 12th JVT meeting: Redmond, Washington, United States, July, 2004.
• Woo-Shik Kim, Dmitry Birinov, and Dae-Sung Cho, Hyun Mun Kim (Multimedia Lab, Samsung AIT), "Enhancements to RGB coding in H.264/MPEG-4 AVC FRExt". ITU-T SG16 Q.6, document VCEG-Z16, 26th VCEG meeting: Busan, Korea, April 2005.
• P. Agawane and K. R. Rao (Multimedia Processing Lab, University of Arlington), "Implementation and evaluation of residual color transform for 4:4:4 lossless RGB coding". International Conference on Recent Advances in Communication Engineering, Hyderabad, India, December, 2008.
• T. Strutz; A. Leipnitz: "Reversible Colour Spaces without Increased Bit Depth and Their Adaptive Selection." IEEE Signal Processing Letters, Vol.22, No.9, 2015, 1269 - 1273
• T. Strutz: "Multiplierless Reversible Colour Transforms and their Automatic Selection for Image Data Compression." IEEE Transactions on Circuits and Systems for Video Technology, Vol.23, No.7, July 2013, 1249-1259

References

1. "Dirac Specification" (PDF). BBC. p. 136. Archived from the original (pdf) on 2015-05-03. Retrieved 2010-05-04.
2. "YCoCg: A Color Space with RGB Reversibility" (ppt). University of Texas at Arlington. Retrieved 2010-05-02.
3. Yair Moshe. "H.264 Amendment: Fidelity Range Extensions" (pdf). Signal and Image processing Lab (SIPL). p. 15. Retrieved 2010-05-02.