Readit NewsIf you want to experiment thatwith your photos I'm available on DM: twitter.com/_dulacp
No wonder we have designers, developers, marketers, product managers etc. Proves to show that you can have the best/most amazing product in the world BUT if you can't communicate what it does, it will always fail.
(this is due to how this model learns, maybe a future model will alleviate that)
> All images produced are released under the CreativeML Open RAIL-M license.
How does this work, legally?
Are the rights to the likenesses of the person in these output images being licensed by someone under the CreativeML Open RAIL-M license?
If so, to whom is it licensed? And by whom?
I guess I should rephrase to something like "The rights of the images are transferred to the user who purchased the service" or hire a lawyer to have it rephrased properly.
$10 for a "well, maybe the computer will generate something good enough that you like, sorry no refunds" is a big turnoff though.
Or a refund policy could seal the deal? :)
It would be great if you could adapt the output to be more realistic.
I'm personally not a fan of beautifying our pictures.
If you are interested in trying, we can talk in DM: twitter.com/_dulacp
Allowing some flexibility in the types of costumes might be tricky though (today it's easy because I only allow for a few outfits for "business" looks).
So now even our profile pictures aren't really us, but just some pseudo-reality version of who we think we are. And I know I know, people will argue that makeup/airbrushing/photoshop/facetune has been going on forever, but at some point I feel like we cross the line where it's no longer "reality with some touchups", but instead it's "complete fantasy made to mimic reality".
I just feel like AI is shooting us head first down this state where reality and fantasy are evermore difficult to differentiate, and I don't like the implications.
It all started when we started to argue with my family members on which avatars looked more like me. It made me realize that we were much more sensitive than I thought about our self-image. Me and my partner would pick different pictures in a set of 10 samples ^^ as if we had two slightly different perceptions of reality.
Now, I changed my mind slightly and tend to see these models as an another type of compression of information. Almost like a new censor of data.
For example, Astria has its own trick to increase the number of headshot in frame, if you download the checkpoint and run it locally you will see that it doesn't handle well negative prompts (which, IMHO, is usually a side effect observed on fine-tuned style models with dreambooth). With this trick you can get up to 70% of headshots in frame but won't reach 100% without risking a loss of prior knowledge. This is my experience so far, maybe you have another opinion based on your experience.