KRЯRL DЯAWINGS: August 31, 2021 -- Figure drawings into Architecture

The Tuesday Night drawing group of Santa Fe drew together online using Zoom, because of the Covid 19 quarantine. The model was from the Washington DC area.

Krrrl Drawings Playlist

KRRRL Figure drawings

into Architecture

John Tollett's drawings from the same session

FIGURE DRAWINGS

into ARCHITECTURE

I used the prompts in the "text-to-image" program VQGAN+CLIP to convert all my pencil figure drawings into architecture:

This pencil figure drawing

was the base initial image

YANDEX and Google similar visual images

prompts: "In the style of unusual architecture"

Paint 3D

PNG

YANDEX and Google visually similar images

Deep Dream Generator alteration

YANDEX and Google similar visual images

prompts: "Adobe architecture"

*****

Base figure drawing, initial image

YANDEX and Google similar visual images

prompts: "In the style of blocky architecture"

YANDEX and Google similar visual images

prompts: "Unusual architecture in the style of Giorgio de Chirico"

Deep Dream Generator alteration

*****

Base figure drawing, initial image

YANDEX and Google similar visual images

prompts: "In the style of bizarre architecture"

Paint 3D

PNG

ARCHITECTURE Summary

PORTRAIT Drawing

AI Buffett

I distorted my portrait drawing through four online AI (artificial intelligence) programs:

The scanned pencil drawing portrait

KRRRL Portrait

passed through AI variations

NOTE: The "Portrait" prompt in VQGAN+CLIP program delivered a very similar image result twice, seen both at the beginning and end of the long video above -- even though the "initial_image" was very different for each run. It seems that the "text-to-image" input overpowers the "initial image" input.

First and last VQGAN images in the above video,

both using the prompt "Portrait" --

note how the results converged on the same colored portrait,

even though they were based on very different initial images

AI (artificial intelligence) online programs used in the portrait video:

All these guys below are related in Artbreeder (and loosely based on my pencil drawn portrait):

Artbreeder

This face below is what comes up when I uploaded my portrait drawing into Deep Face Drawing:

Deep Face Drawing

Deep Dream Generator alteration

PEN DRAWINGS

I colored my pen drawings from that session in Petalica Paint:

(ok, this is a colored pencil drawing)

I then created a few PNG images with transparent backgrounds in Paint 3D:

COMPOSITIONS

The compositions were created in Photoshop Elements 2021 using "Layers."

YANDEX and Google visually similar images

PNG

Paint 3D

Composition in 3D

RELIEF PRINTING

I drew the last pose on linoleum, and then immediately cut it out and printed it with the Tortilla Press:

Relief Print

UPDATE (October 1, 2021): I submitted an edition of 45 of this relief print for the Southwest Print Fiesta Print Exchange 2021, all made on the Tortilla Press.

"Abstract Reclining Figure"

Edition of 45 prints

for the Southwest Print Fiesta Print Exchange 2021

The print exchange

exhibited at the Southwest Print Fiesta 2021

COMPARING VQGAN Programs

I used the same "secret sauce" text that Holly used to create her Abraham.ai image. I used it in the VQGAN+CLIP Colab notebook, starting with my portrait as the "initial_image," and the result was similar, though different enough:

The above two images combined sorta,

to make the large bottom image --

created with the same text,

but in different VQGAN programs

RIOT

3000 x 5178 pixels

but only 231.70 KB,

reduced by RIOT

Pete McCracken from Portland gave a silk screen presentation at Remarque Print Workshop in Albuquerque on September 2nd. He is giving a Serigraph Print Editions workshop at Remarque on November 6 - 7, 2021. It would be nice to pull an AI print edition, and bring the new found digital art into the real world.