Adding to Cart…
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.You currently have no notifications.
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
...so I had to do it.
NVIATWAS (using "Cinematic" art style )
Unfortunately cannot post the results.
...yes indeed, like the SUV, this isn't going to be a passing fad. it will be a boon to production companies and co course a threat to those who work in them. Witness the writers and actors strikes of last summer which forced film and video streaming production firms to reconsider turning completely to this technology.. .
One word I noticed in the video which this stuck with me most pretty much all my life is "craftsmanship". That is something which will never disappear. People didn't out down pencils pens and paintbrushes when the camera was introduced. Similarly so with the current generation of digital art. I have no issue with AI being part of a tool or process to make it more accurate (like denoising). Where I draw the line is having algorithms generate what is supposed to be "creative" art be it an image, a novel or music composition. (add digital synthesis to the latter and who needs anyone to perform the work an instrument?).. Generative AI is taking the creative human element out of the equation.
Yes I transitioned from traditional art media to 3D CG but it wasn't so much that it was the next cool thing, it was because crippling arthritis made it more and more difficult to use traditional tools. It has already taken away music (classical keyboard another pursuit I spent many years in) At least 3D let me stay in touch with one of the thing's I loved and it stil took a sense of craft to create custom or personalised characters and scenes.
Last night I played around with DA's Dream Up AI engine and found it little more than a toy for amusement rather than something actually creative. I jokingly had it do an NVIATWAS image entering the full words in the parameters field, (when I used just the acronym it seemed pretty confused) and yes it basically followed the parameters and generated a scene, but it wasn't quite what I had envisioned in my mind. Even using lengthy and detailed parameters along with setting the threshold higher, every time I felt it was pretty much a "crap shoot" as to what the final image would actually look like. Again I'd have one image in my head but the result was nothing like what I expected. Now true their process is more limited by what they use for the learning base. IN the end none of those images were "mine" as they were the result of what an algorithm mixed up based on the parameters I entered.
In a way I feel more like it is "digital collage" as it takes the terms entered then draws from what is loaded in it's learning base memory and blends something together that basically tried to match the instructions given.
Again, an amusing toy at best not quite up to paint by numbers yet as that still took a modicum of actual skill. All I can say is thankfully I am retired and do 3d art for relaxation as well as feel a sense of personal purpose rather than for a living..