A very complete and simple script to resize, combine/mix 2 luts or convert your CLUT images to CUBE format and viceversa.

Do you like this project? Support it by donating

• Paypal: Donate or paypal.me/michelerenzullo

• Bitcoin: 1K9RF3s4aocmaRbh2Zu2FuHjrcg5BNeDxU

• Auto-detect input format

• Auto-resize not perfect square LUT size(example #33, #17...)

• Flip RGB values

• Change size and shape of your LUT (example Spark AR LUTs to Lens Studio 1x16 LUTs)

• Generate identity CLUT

• Read all types of CLUT

• Combine 2 luts and specify the amount from lut1 (0) to lut2 (100)

• Two type of interpolation when resizing: Tetrahedral or Nearest

Square:

Square one-row:

Hald:

• Python 3

• Numpy

• Pillow

$ pip install numpy pillow

$ LUTify.py -h

usage: LUTify.py [-h] [--input INPUT] [--combine COMBINE] [--mixer MIXER]
[ --preserve] [--output OUTPUT] [--format {hald,square}] 
[--identity] [--size SIZE] [--method {nearest,tetrahedral}]
[--rows ROWS] [--flip]

...

• -i INPUT/-o OUTPUTsupports .CUBE, .PNG, .JPG, .TIFF
• -c COMBINE/ supports .CUBE, .PNG, .JPG, .TIFF, optionals:

• • -p PRESERVE when combining 2 luts preserve the max size between them, default false
• • -x MIXER mix amount from lut1(0%) to lut2(100%), default is 50

• -id IDENTITY generate a CLUT identity

• -f FORMAT "hald" or "square", override default output:

• • if output is CUBE or "identity", default format is "hald"
• • if input is CUBE and output is image the default format is "hald"
• • if input is image and output is image default format will be automatically determinated as the opposite of the input format.

• -s SIZE choose your CLUT size or LUT size overriding input value.

• -m METHOD the method of interpolation when resizing between "tetrahedral"(default) and "nearest"

• -r ROWS number of rows when output is square

• -ud FLIP flip upside down RGB values

If you would like to convert your Spark AR LUTs to Lens Studio:

$ LUTify.py -i "SparkAR_SQUARE.png" -o "LensStudio_SQUARE.png" -s 4 -r 1 -f square --flip