AI Colorization

With the Colorization effect it is possible to automatically recolor black and white film material. An AI-supported algorithm determines the most likely colors within the two color dimensions red green and blue yellow.


Black and white	Colorized

Luminance enhancement	If the image is better exposed, the color prediction works better, so you can optimize the brightness. You can also specify whether the enhanced luminance should be applied to the material (Apply to output) or only used to improve color prediction (Estimation only - Do not apply to output).
Precision	The precision of the color application can be defined in four different levels: Low, Medium, High, or Match input size. The latter uses the full resolution of the original image for color prediction, the others use a more or less downscaled version of the image.

The following parameters are used to adjust the overall coloring of the image. All parameters can be automated in order to react to changing light conditions in the film material.

Invert Green and Red	The assignment of determined color attributes in the red-green channel is swapped.
Invert Blue and Yellow	The assignment of determined color attributes in the blue-yellow channel is swapped.
Shift between Green and Red	The colors in the Red-Green channel are shifted towards red or green.
Shift between Blue and Yellow	The colors in the Blue-Yellow channel are shifted towards blue or yellow.
Gain on Green and Red	Control of the strength of the colorization in the red-green channel.
Gain on Blue and Yellow	Control of the strength of the colorization in the blue-yellow channel.