| Author | Thread |
|
|
01/22/2025 11:29:12 PM · #51 |
Challenge: Backsides II
Title: Girl Hiker in a Fantasy Forest
AI: ChatGPT
Message edited by author 2025-01-22 23:39:11. |
|
|
|
01/22/2025 11:38:38 PM · #52 |
Challenge: Backsides II
Title: Girl Hiker in a Fantasy Forest II
AI: Midjourney Raw Model |
|
|
|
01/22/2025 11:42:32 PM · #53 |
| I am working on an idea for a film-making project in Unreal Engine 3D. The little girl hiker would be the character, I am hoping that AI will help me refine ideas. |
|
|
|
01/23/2025 02:51:10 AM · #54 |
Originally posted by GeneralE:
I'm not using any AI stuff right now, but maybe one of you bird fanciers might want to have a go at this picture of mine ...
 |
I tried it in Photoshop and all I got was pictures of pigeons despite using the prompt "bird with the body of a pigeon and the head of a dog". Maybe someone can get it to work with other software. |
|
|
|
01/23/2025 02:52:11 AM · #55 |
Originally posted by MargaretNet:
Challenge: Backsides II
Title: Girl Hiker in a Fantasy Forest
AI: ChatGPT |
I love this one. |
|
|
|
01/23/2025 05:03:13 AM · #56 |
Originally posted by GeneralE:
I'm not using any AI stuff right now, but maybe one of you bird fanciers might want to have a go at this picture of mine ...
|
Took quite some time, ChatGPT could not produce an image but created a prompt for Midjourney which I still needed to modify to get anything close to your image. |
|
|
|
01/23/2025 05:09:27 AM · #57 |
Originally posted by GinaRothfels:
Originally posted by MargaretNet:
Challenge: Backsides II
Title: Girl Hiker in a Fantasy Forest
AI: ChatGPT |
I love this one. |
Thanks, I prefer it as well as it will fit better with 3D technology. |
|
|
|
01/23/2025 06:07:22 AM · #58 |
Originally posted by MargaretNet:
Originally posted by GeneralE:
I'm not using any AI stuff right now, but maybe one of you bird fanciers might want to have a go at this picture of mine ...
|
Took quite some time, ChatGPT could not produce an image but created a prompt for Midjourney which I still needed to modify to get anything close to your image. |
Well done. I tried your prompt in Photoshop and got an interesting image, but it's still a pigeon with no sign of a dog.
 |
|
|
|
01/23/2025 10:00:07 AM · #59 |
| Wow, that's a creature! Eerily strange. Margaret's is cool also, but way cuter :-) |
|
|
|
01/23/2025 10:43:24 AM · #60 |
add asymmetry to the list of things ChatGPT doesn't understand. Thank goodness Dear Leader will be using our taxes to subsidize its continued improvement.
You said:
create a photo that is asymmetrical but has a centered composition.
ChatGPT said:
Here's an image that embodies an asymmetrical yet centered composition. Let me know if you'd like any adjustments or refinements!
You said:
Please make it more photographic and more asymmetrical
ChatGPT said:
Here's a more photographic and asymmetrical version of the scene, while keeping the centered composition intact. Let me know your thoughts or if you'd like further adjustments!
You said:
it's more photographic but it's very symmetrical. Please make it not be symmetrical at all.
ChatGPT said:
Here is an image that emphasizes asymmetry throughout, while maintaining a natural, photographic style. Let me know if this captures what you had in mind or if you'd like further refinements!
You said:
you have no idea what asymmetry is. all you give me is symmetry. I don't want symmetry.
ChatGPT said:
Thank you for your feedback—I understand your frustration. Let me address this properly and deliver something that truly embodies asymmetry. I'll focus on creating a design where balance is not achieved through symmetry but through the thoughtful arrangement of uneven elements. Let me try again.
Here's an image that leans fully into asymmetry, with uneven elements and an off-center focal point. Let me know if this better matches your vision or if you'd like me to refine it further!
Message edited by author 2025-01-23 10:43:54. |
|
|
|
01/23/2025 10:51:40 AM · #61 |
Originally posted by posthumous:
add asymmetry to the list of things ChatGPT doesn't understand. Thank goodness Dear Leader will be using our taxes to subsidize its continued improvement.
You said:
create a photo that is asymmetrical but has a centered composition. |
Don, can you show us an actual image that matches your criteria? My brain is slowing down but I'm having a hard time visualizing "centered-but-asymmetrical"...
ETA: aside from the obvious asymmetrical object centered on a featureless ground...
Message edited by author 2025-01-23 10:54:04. |
|
|
|
01/23/2025 11:10:51 AM · #62 |
Using chatGPT CALL.E engine.
It shows that it really helps to provide a description of surroundings, atmosphere etc.
The prompts are shown underneath the images
Message edited by author 2025-01-23 11:11:31. |
|
|
|
01/23/2025 11:20:05 AM · #63 |
Originally posted by Bear_Music:
Originally posted by posthumous:
add asymmetry to the list of things ChatGPT doesn't understand. Thank goodness Dear Leader will be using our taxes to subsidize its continued improvement.
You said:
create a photo that is asymmetrical but has a centered composition. |
Don, can you show us an actual image that matches your criteria? My brain is slowing down but I'm having a hard time visualizing "centered-but-asymmetrical"...
ETA: aside from the obvious asymmetrical object centered on a featureless ground... |
https://en.m.wikipedia.org/wiki/Time_Transfixed#/media/File%3ATime_transfixed.jpg
https://www.wikiart.org/en/max-ernst/the-virgin-spanking-the-christ-child-before-three-witnesses-andre-breton-paul-eluard-and-the-1926/
https://www.manray.net/assets/img/man-ray-photo.jpg
 
Message edited by author 2025-01-23 14:58:07. |
|
|
|
01/23/2025 02:52:29 PM · #64 |
Originally posted by MargaretNet:
Originally posted by GeneralE:
I'm not using any AI stuff right now, but maybe one of you bird fanciers might want to have a go at this picture of mine ...
|
Took quite some time, ChatGPT could not produce an image but created a prompt for Midjourney which I still needed to modify to get anything close to your image. |
Thanks, that looks great -- I wish I could remember how long it took me to make the original.
Just wondering, what resolution/image size do these AI engines typically produce? |
|
|
|
01/23/2025 02:55:16 PM · #65 |
Originally posted by GinaRothfels:
Originally posted by MargaretNet:
Originally posted by GeneralE:
I'm not using any AI stuff right now, but maybe one of you bird fanciers might want to have a go at this picture of mine ...
|
Took quite some time, ChatGPT could not produce an image but created a prompt for Midjourney which I still needed to modify to get anything close to your image. |
Well done. I tried your prompt in Photoshop and got an interesting image, but it's still a pigeon with no sign of a dog.
|
What Bear said -- it's a great picture, even if not what you asked for.
Thanks for experimenting! |
|
|
|
01/23/2025 11:59:36 PM · #66 |
Originally posted by GeneralE:
Just wondering, what resolution/image size do these AI engines typically produce? |
I did not know so I asked ChatGPT for the two generators I used:
Midjourney
Single-Pass Max: ~1024×1024 by default.
Upscale to Max: Up to ~2048×2048 (1:1) or equivalent pixel counts for other aspect ratios.
Further Increases: Done through multi-step stitching or the “Zoom Out” feature.
DALL·E 2
Single-Pass Max: 1024×1024.
Outpainting: You can keep adding 1024×1024 tiles to expand the image in any direction. Effectively, there’s no strict upper limit on total resolution (it is limited only by how many outpainting steps you want to perform). |
|
|
|
01/24/2025 01:49:06 PM · #67 |
| Thanks -- that kinda what I expected (images suitably sized for the internet but not much else). |
|
|
|
01/26/2025 06:03:54 AM · #68 |
Reference to my entry
Using chatGPT DALL.E engine
and prompt:
A photorealistic image of the town Corte on Corsica, situated on the hill, with the mountain as backdrop and the low evening sun coming from the left, with close-by foliage in the lower part of the image.
and then improved by the additional prompt
The view should be from a low standpoint looking up towards the old village.
 |
|
|
|
02/07/2025 01:50:15 AM · #69 |
It doesn't look that this is going to be ongoing. Maybe a scored AI challenge would give more motivation and participation? As proof of ownership the participants could submit screen captures of their prompts and generated images.
I cancelled Midjourney as I prefer ChatGPT to create images for me, it is more interactive and intuitive although it is still difficult to keep the same character and style that I need for my 3D movie. |
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 07/31/2025 04:51:23 PM EDT.
