Colour (Feb 2024) + Wombo AI versions by 3dcheapskate
Close
Hover over top left image to zoom.
Click anywhere to exit.
Members remain the original copyright holder in all their materials here at Renderosity. Use of any of their material inconsistent with the terms and conditions set forth is prohibited and is considered an infringement of the copyrights of the respective holders unless specially stated otherwise.
Description
(An old render)
This was the original image from which the monochrome one ultimately came.
1 Mar 2024 - Let's see what Wombo can do with my render
I thought I'd try something slightly different this time - setting the influence of my input image to high.
Image generator: Dream by WOMBO - the free online version at dream.ai, not the app.
Prompt: "girl in lace teddy"
Style: Left to right: Baroque v3, Cartoonist v3, Comic v3, Dark Fantasy v3, HDR v3, Ink v3, Sketch v3
Input image: My render, i.e. the main image here. Influence set to strong
My notes: Most presses of the 'Make Art' button produced similar results regardless of the (v3) style I selected. I couldn't really tell which picture was which style unless I'd noted it, except for Baroque v3 which has a unique slant on faces.
My favourites: I think they're all much of a muchness - very close to my original image, but that unmistakeable Poser-renderness has gone and there's a more painterly feel.
Leaving the influence at normal, the default setting, leads to a much greater variety and it's quite easy to spot which style is being used.I usually need to add a few things to the prompt to make sure it doesn't stray too far from my render. Details for the ninth to fifteenth images (eighth to fourteenth Wombos) below:
Image generator: Dream by WOMBO - the free online version at dream.ai, not the app.
Prompt: "young woman in lace teddy, hands behind her back, head turned slightly left, hair covering eye"
Style: Left to right: Baroque v3, Cartoonist v3, Comic v3, Dark Fantasy v3, HDR v3, Ink v3, Sketch v3
Input image: My render, i.e. the main image here. Influence set back to normal
P.S. If you'd like to try it out yourself you're welcome to use any of these pictures as an input image to whatever text+image-to-image software you like.
My reference(s):
Pinup04 purple 2C.pz3
Comments (5)
FirstBastion
This was a interesting experiment. Given how nicely lit the original was, the Ai had lots of quality to work with. The two that stand out are the Rubenesque and the decidely Anime
3dcheapskate
I've been experimenting a lot as you'll see if you search my gallery for "wombified", a tag I now use for any upload that includes both my original Poser render and Wombo Dream image&text-to-image versions. I've been surprised at how well* (sometimes) it extracts the poses from my renders, even in some of the multi-figure renders with lots of complexity (e.g. search my gallery for the wombified "snake dance")
When I was doing this particular set of tests I found myself wishing that Wombo Dream had an input image influences between normal and strong. I found all the 'nrmal' Wombos too samey, and all the 'strong' Wombos too stylishly distinct. However, with the next set of images I've uploaded ("My best Wombified Poser render so far ...") I think I'm beginning to find a way forward there... (and with "Babe again - still not the piglet" I think I managed to get Baroque v3 to just tweak a previous Wombo and work some classic-oil-paintingly magic...)
*sometimes... definitely not always! Although the failures are often rather enlightening as I can usually look at my render and see the reason for the misinterpretation.
brain1969
great work on the light and good pose for it
3dcheapskate
Thanks :o) The lighting and camera viewpoint were inspired by a picture I saw at DeviantArt - my original attempt is posted in my gallery her as "2024-01-11 Inspired by a picture at DeviantArt".
VirtualCity
Freaking Awesome! I love the lighting and emotion of this piece. I also like how you posted several other renditions of the original for people to generate ideas from. Very talented work. Thanks for all the information on how you did all this. It makes me consider adding an Image generator to my pallet after I make a render in Poser.
3dcheapskate
I'd definitely recommend giving it a try. If you can find a render with a single figure and little or no background that's probably a good first test, along with a simple-as-possible prompt of just one or two words.
NightCafe's not doing it for me yet, but with Wombo Dream and its latest v3 styles I think I'm personally on the right track.
I'd just recommend playing, having fun, and if anything seems to work for you just go with the flow. If it doesn't then forget about it... for a few months... ;o)
water
Super !
3dcheapskate
Thanks :)
PhthaloBlue
Beautiful!
3dcheapskate
Thanks :)