-
Sign up or login, and you'll have full access to opportunities of forum.
Art by Art
- Thread starter Art
- Start date
-
- Tags
- oiginal art
Art
Governor
The good old lady is in a good mood. She certainly has a lot of fun doing what she does.
That on the subject of good sex at an old age.
Attachments
Art
Governor
Here, as already announced, several AI generated images.
I generated these images to get to know Tensor.Art. So far, many images have turned out differently than expected, despite a clear prompt.
I probably still need to practice a lot. This is also necessary, because there are many possibilities in the prompt and LoRa definition. But what fascinated me most was the interpretation of the AI to my specifications. There were always surprises in the results.
However, I will not continue this kind of image generation. If, then only for my private archive.
In the future, I will concentrate on reworking my 3D images with AI and, I hope, improving them.
The following images are posted without further comment.
One more word,
If someone wants to have a cropped image from the selection of images, for manips, just name the number in a comment. I will then publish this image as PNG with transparent background.
I generated these images to get to know Tensor.Art. So far, many images have turned out differently than expected, despite a clear prompt.
I probably still need to practice a lot. This is also necessary, because there are many possibilities in the prompt and LoRa definition. But what fascinated me most was the interpretation of the AI to my specifications. There were always surprises in the results.
However, I will not continue this kind of image generation. If, then only for my private archive.
In the future, I will concentrate on reworking my 3D images with AI and, I hope, improving them.
The following images are posted without further comment.
One more word,
If someone wants to have a cropped image from the selection of images, for manips, just name the number in a comment. I will then publish this image as PNG with transparent background.
Attachments
Art
Governor
I did a photo manipulation just for fun with an AI generated image and a scene photo from a Cruel video.
The AI image with green background and the figure cropped from it I show here together with the result. It's not my genre, but it's fun to think outside the box sometimes.
The AI image with green background and the figure cropped from it I show here together with the result. It's not my genre, but it's fun to think outside the box sometimes.
Attachments
Art
Governor
Beauty is in the eye of the beholder. Everyone has his own taste.
Me!
Tensor.Art is available to everyone for free. You show it to us.
hurdygurdy57
Magistrate
M & Thx
Executioner
I love what you have done with your AI! I think the women are real looking and are horrified for their circumstances.Can you do an AI series using me being crucified?
I’ll send you some pictures. I would love to see what I would look like crucified mmmmm
Thank you so much!
M
hurdygurdy57
Magistrate
Hangnail
Tribune
I agree 100% on both points.
chimaera
Magistrate
I don't really know what I am doing with AI, and have had a lot of issues with skin looking plastic or supermodel flawless- are you having the same?. In Tensor I have had better result with the 'legacy' SD 1.5 than the SDXL, even though the latter is supposed to be much better. The best results I have had are with the Model as either EpiCRealism-pure Evo or Realism Fusion and the LoRAs two out of Add More Details, Skin and Hands (male/female) and/or EpiCRealism Helper. Also I put "cartoon, comic" in the negative prompt and "digital photo" in the prompt. I am sure there are hundreds of other combinations, but these seem to work. Often EpiCRealism-pure Evo seems to be better for indoor/gloomy images.
BTW, a range of my experiments are at https://www.deviantart.com/harriotta/gallery/88819840/captives-and-dungeons
These have used pretty much every free online platform I could found, and some have had a workflow where I did a basic image in one and improved it in another. You can see some are not bad, some pretty poor.
BTW, a range of my experiments are at https://www.deviantart.com/harriotta/gallery/88819840/captives-and-dungeons
These have used pretty much every free online platform I could found, and some have had a workflow where I did a basic image in one and improved it in another. You can see some are not bad, some pretty poor.
Last edited:
PunisherX
Assistant executioner
It's so unrealistic. As much the pics are technically realistic as much the potential victims sre unrealistic. No torturer with free choice would choose an ugly old woman for torture, while the beauties are waiting in the dungeon... It's the same like women torture women - unrealistic. I'm sorry, as long as there are only ugly old ones with ugly bodies on the focus i'm simply out and have to look elsewhere. Good luck.
Art
Governor
I wish you good luck too!
You are not forced to look at pictures which you don't like.
Just do not take it personally, but you should understand that I create the images under my own direction and with not inconsiderable costs. That's why I mainly decide what I show and how the people in my pictures look like. It is my hobby.
Comments and likes to the pictures are my reward. The counter shows me the success of each picture.
Even if it is not visible to you, I appreciate your comments. Because every opinion is important to me.
Greetings
Artur
Art
Governor
Too bad, unfortunately I'm not there yet with my AI skills. The AI doesn't even recognize simple torture tools at the moment. How should I then manage a solid crucifixion. Maybe later, if I continue the topic of AI.
I'm still practicing dealing with what the AI is currently offering me. Look my later images.
Art
Governor
DA I can't see at the moment, they kicked me out of there a long time ago as one of the first. I have to activate my replacement figure there. But I have seen your AI pictures on Pixiv. They are great, but also very well-behaved, considering your 3D pictures.
You are much more advanced than me, though. So far, I have not tried to display two people. Well, I've only been doing it for a few days.
I'll just attach a screenshot of my latest settings here.
Attachments
Art
Governor
Fifty Blows.
About the creation of the image.
The basic image, Woman - Bank - Room, was created with the AI program, Tendor.Art. I set this AI image as the background image in DazStudio and placed the virtual soldier in front of the background image, then rendered it together.
In a comic program, COMIC LIVE 3, I then added the text and speech bubbles.
In PhotoScape, a photo program, I then cropped the image format and added my signature.
I covered the slightly misshapen hand with the text box, and the ribs also look a bit unnatural. I do not retouch such things, it should be recognized that the basis is an AI image. But I think that the result is quite decent.
It is also strictly speaking not a real photo manipulation. It is the process of creating an image with non-existent people, i.e. AI photo with the image of a woman and a virtual 3D soldier from a 3D program.
In the future, it must be decided at a higher level whether and how AI images, in the publication, must or should be clearly marked.
About the creation of the image.
The basic image, Woman - Bank - Room, was created with the AI program, Tendor.Art. I set this AI image as the background image in DazStudio and placed the virtual soldier in front of the background image, then rendered it together.
In a comic program, COMIC LIVE 3, I then added the text and speech bubbles.
In PhotoScape, a photo program, I then cropped the image format and added my signature.
I covered the slightly misshapen hand with the text box, and the ribs also look a bit unnatural. I do not retouch such things, it should be recognized that the basis is an AI image. But I think that the result is quite decent.
It is also strictly speaking not a real photo manipulation. It is the process of creating an image with non-existent people, i.e. AI photo with the image of a woman and a virtual 3D soldier from a 3D program.
In the future, it must be decided at a higher level whether and how AI images, in the publication, must or should be clearly marked.
