Lecture 20: Introduction to Color Science (Cont) (68)
GKohavi
Here's a pretty interesting paper that tries to predict the a, b channels from the L channel. What's interesting is that they also show how having non-uniformity in the ab histogram can cause some problems when predicting. https://arxiv.org/abs/1603.08511
andrewdcampbell
The LAB color space is commonly used as an intermediate when graphics for print have to be converted from RGB to CMYK (which most printers use), as the CIELAB gamut includes both the gamuts of the RGB and CMYK color models.
amandaawan
Unlike the RGB and CMYK color models, the CIELAB colors are designed to approximate human vision. We can change the color balances with a* and b* (cones) and brightness using L*(rods). Because it has a larger color space than RGB and CMYK, in the past, with 8-bit channels/bitmaps, converting RBG into CIELAB then back into RGB was a lossy operation. Now that we use 16-bit, the loss is negligible.
Here's a pretty interesting paper that tries to predict the a, b channels from the L channel. What's interesting is that they also show how having non-uniformity in the ab histogram can cause some problems when predicting. https://arxiv.org/abs/1603.08511
The LAB color space is commonly used as an intermediate when graphics for print have to be converted from RGB to CMYK (which most printers use), as the CIELAB gamut includes both the gamuts of the RGB and CMYK color models.
Unlike the RGB and CMYK color models, the CIELAB colors are designed to approximate human vision. We can change the color balances with a* and b* (cones) and brightness using L*(rods). Because it has a larger color space than RGB and CMYK, in the past, with 8-bit channels/bitmaps, converting RBG into CIELAB then back into RGB was a lossy operation. Now that we use 16-bit, the loss is negligible.