|
|
|||||||||
|
|||||||||
| |||||||||
|
|
|
| |||||||||
![]() |
|
|
«
Previous Thread
|
Next Thread
»
|
Thread Tools | Search this Thread | Display Modes |
|
#1
|
|||
|
|||
|
Better Chroma Subsampling
1- Chroma subsampling creates out of gamut values say, when non-
neutral color difference components are reconstructed onto black pixels. The article gives one example (red text on a black background). This could be avoided with in-range chroma reconstruction. The resulting Y'CbCr values can lie outside R'G'B' gamut. That's what I mean by out of gamut. 2- Regarding constant luminance, I define luma and luminance differently. Luminance takes into account gamma correction whereas luma is just an approximation (which works well enough). It's explained in the article. 3- It's not entirely necessary to specify how the chroma samples are interpolated. It's ok to leave it up to the output device to choose the interpolation method based on tradeoffs between quality and cost. (Though in some other applications like studio video, the rec. 709 standards has filter guidelines so that presumably multi-generation performance will be good.) Glenn |
![]() |
| Viewing: Web Development Archives > FAQs > Compression > Better Chroma Subsampling |
| Thread Tools | Search this Thread |
| Display Modes | Rate This Thread |
|
|
|
|