Outcomes created using filter OPUS VERMICULATUM VITREUM (2023)

[dinasset]

Third image.

-colors as they are

2023-09-19 06-10-38 young-woman-1867618_1920, as a Roman Mosaic (parms=8,0,0,3,0,0,2,0).jpg (646.5 KiB) Viewed 845 times

-colors stretched

2023-09-19 06-07-41 young-woman-1867618_1920, as a Roman Mosaic (parms=8,1,0,3,0,0,2,0).jpg (653.56 KiB) Viewed 845 times

-colors equalized

2023-09-19 06-09-04 young-woman-1867618_1920, as a Roman Mosaic (parms=8,2,0,3,0,0,2,0).jpg (674.39 KiB) Viewed 845 times

[dinasset]

Fourth image.

-colors as they are

2023-09-19 06-04-19 woman-1509959_1920, as a Roman Mosaic (parms=8,0,0,3,0,0,2,0).jpg (688.54 KiB) Viewed 845 times

-colors stretched

2023-09-19 06-05-43 woman-1509959_1920, as a Roman Mosaic (parms=8,1,0,3,0,0,2,0).jpg (685.08 KiB) Viewed 845 times

-colors equalized

2023-09-19 06-02-56 woman-1509959_1920, as a Roman Mosaic (parms=8,2,0,3,0,0,2,0).jpg (647.39 KiB) Viewed 845 times

[dinasset]

My bottom line?
If you care about a portrait, try all the 3 options to choose among them the one you like better.

[dinasset]

Did some experiments to generate shining mosaics.

[dinasset]

First triplet.

2023-09-19 13-53-58 couple-1854081_1920, as a Roman Mosaic (parms=12,2,1,3,4,7,2,2).jpeg (709.73 KiB) Viewed 836 times

2023-09-19 13-52-05 couple-114328_1920, as a Roman Mosaic (parms=12,2,1,3,4,7,2,2).jpeg (497.65 KiB) Viewed 836 times

2023-09-19 13-50-28 couple-6548045_1920, as a Roman Mosaic (parms=12,2,1,3,4,7,2,2).jpeg (461.96 KiB) Viewed 836 times

[dinasset]

...and second one.

2023-09-19 13-49-07 couple-1850073_1920, as a Roman Mosaic (parms=12,2,1,3,4,7,2,2).jpeg (521.51 KiB) Viewed 836 times

2023-09-19 13-46-52 par-1694334_1920, as a Roman Mosaic (parms=12,2,1,3,4,7,2,2).jpeg (560.66 KiB) Viewed 836 times

2023-09-19 13-44-15 couple-1822585_1920, as a Roman Mosaic (parms=12,2,1,3,4,7,2,2).jpeg (586.05 KiB) Viewed 836 times

[dinasset]

The best one IMO is the one using image 114328, but of course my evaluation is biased by my age....

[dinasset]

This afternoon I dedicated my time in producing mosaics of cityscapes.
Alas, they are su full of details that I needed to use the minimum tilesize (1), and the time to render the mosaic was looooong.

[dinasset]

I show here the oucome both without and with the option EQUALIZE (you know already I'm a fan of those unexpected colours...).

First image:

-without equalize (i.e.original image colours)

2023-09-19 16-28-25 panorama-2646143_1920, as a Roman Mosaic (parms=20,0,0,1,0,0,1,0).jpeg (2.27 MiB) Viewed 826 times

-with equalize preproc

2023-09-19 16-57-39 panorama-2646143_1920, as a Roman Mosaic (parms=20,2,0,1,0,0,1,0).jpeg (2.36 MiB) Viewed 826 times

not much different

[dinasset]

Secon image.

original colours

2023-09-19 16-32-13 buildings-2602324_1920, as a Roman Mosaic (parms=20,0,0,1,0,0,1,0).jpeg (2.47 MiB) Viewed 826 times

equalized

2023-09-19 17-02-09 buildings-2602324_1920, as a Roman Mosaic (parms=20,2,0,1,0,0,1,0).jpeg (2.51 MiB) Viewed 826 times

differences more evident

[dinasset]

Third image

original colours

2023-09-19 16-37-31 hamburg-3846525_1920, as a Roman Mosaic (parms=20,0,0,1,0,0,1,0).jpeg (2.65 MiB) Viewed 870 times

equalized

2023-09-19 17-17-25 hamburg-3846525_1920, as a Roman Mosaic (parms=20,2,0,1,0,0,1,0).jpeg (2.54 MiB) Viewed 870 times

clear differences

[dinasset]

Fourth image

original colours

2023-09-19 16-41-45 street-3453557_1920, as a Roman Mosaic (parms=20,0,0,1,0,0,1,0).jpeg (2.89 MiB) Viewed 870 times

equalized

2023-09-19 17-08-49 street-3453557_1920, as a Roman Mosaic (parms=20,2,0,1,0,0,1,0).jpeg (2.72 MiB) Viewed 870 times

clear differences, IMO better equalized

[dinasset]

Fifth (and last) image

original colours

2023-09-19 16-46-36 greece-3384386_1920, as a Roman Mosaic (parms=20,0,0,1,0,0,1,0).jpeg (3.13 MiB) Viewed 870 times

equalized

2023-09-19 17-22-23 greece-3384386_1920, as a Roman Mosaic (parms=20,2,0,1,0,0,1,0).jpeg (3.22 MiB) Viewed 870 times

here I prefer original

[dinasset]

Do you remember post #44 with the problem of colour areas?
Now David updated Posterize bringing the maximum of colours from 32 to 64, hence that problem can now be solved, see:

2023-09-20 05-45-06 schrecksee-2534484_1920, as a Roman Mosaic (parms=64,0,1,3,0,0,0,0).jpeg (831.58 KiB) Viewed 865 times

now we can distinguish very well the mountain from the lake.

Thanks a lot, David

[dinasset]

Tried immediately also che child portrait with 64 colours.
Here it is:

2023-09-20 06-14-15 face-2051719_sq, as a Roman Mosaic (parms=64,0,1,4,0,0,2,3).jpeg (691.96 KiB) Viewed 863 times

I'm very happy.

[dinasset]

A portfolio of portraits.

2023-09-20 06-30-18 girl-1506653_1920, as a Roman Mosaic (parms=64,0,1,4,0,0,2,3).jpeg (765.73 KiB) Viewed 862 times

2023-09-20 06-27-16 woman-3287956_1920, as a Roman Mosaic (parms=64,0,1,4,0,0,2,3).jpeg (711.51 KiB) Viewed 862 times

2023-09-20 06-25-06 woman-1692849_1920, as a Roman Mosaic (parms=64,0,1,4,0,0,2,3).jpeg (856.35 KiB) Viewed 862 times

[dinasset]

...and

2023-09-20 06-21-40 girl-797837_1920, as a Roman Mosaic (parms=64,0,1,4,0,0,2,3).jpeg (774.46 KiB) Viewed 862 times

2023-09-20 06-20-08 woman-1466628_sq, as a Roman Mosaic (parms=64,0,1,4,0,0,2,3).jpeg (688.74 KiB) Viewed 862 times

[dinasset]

Two more (for mine -and hopefully your- pleasure)

2023-09-20 06-37-37 woman-3990680_1920, as a Roman Mosaic (parms=64,0,1,3,0,0,2,3).jpeg (764.6 KiB) Viewed 862 times

2023-09-20 06-36-32 woman-3083453_1920, as a Roman Mosaic (parms=64,0,1,3,0,0,2,3).jpeg (597.01 KiB) Viewed 862 times

[Issabella]

Awesome results, Dinasset.

[Issabella]