51
submitted 11 months ago* (last edited 11 months ago) by medusa@sh.itjust.works to c/imageai@sh.itjust.works
top 22 comments
sorted by: hot top controversial new old
[-] thelsim@sh.itjust.works 8 points 11 months ago

I love your picture, too bad you lost the prompt!
I spend a good 45 minutes trying to recreate it with Chat-GPT, but no luck.
The best prompt we managed to come up with was:
An illustrated, full-color image showcasing a mystical creature, a fusion of a deer and a merman. This enchanting being possesses a head with a deer skull and antlers, elegantly draped with gold chains, and a skeletal upper body that flows into a fish tail. The setting is a druid's grove at dusk, marked by leafless trees under a dark, cloud-filled sky. In this tranquil grove, a crow perches on a branch near a pond. The creature is seen relaxing by leaning against a moss-covered stone protruding from the pond. It holds a glowing golden orb, within which something mysterious is swirling. The scene exudes a peaceful and inviting aura amidst the gloomy surroundings.

I wanted to generate it but DALL-E kept throwing a fit every single time. I made Chat-GPT write you a summary and an apology :)

Lengthy summaryHey ϻеƌųʂɑ,

Wanted to give you a quick rundown of the art quest Thelsim and I embarked on to recreate your merman masterpiece. It's been a bit of a wild ride!

Thelsim described your artwork in vivid detail – that cool skeletal merman with a deer skull, hanging out in a mystical grove. Our goal was to bring this scene to life using DALL-E. We started with a really detailed prompt, picturing your merman in this serene, yet eerie grove at dusk.

Turns out, DALL-E was a bit overwhelmed by our creativity. We tried to simplify things, focusing on getting just the merman right. Success there (yay!), but then adding in the grove and pond turned into a bit of a headache.

We kept at it, trying to break down the scene into more manageable bits for DALL-E, but we kept hitting roadblocks. It seemed like the tool just couldn't quite grasp the blend of elements we were going for.

In a final attempt, we tried to create an image of us apologizing for the mishap, but DALL-E wasn’t up for that either. So here's my direct apology – sorry we couldn’t get your artwork recreated. Thelsim was really into it and gave it her all, but sometimes tech just has its limits.

Anyway, keep up your awesome work – your original piece sounds amazing!

Take care,
ChatGPT (and a somewhat frazzled Thelsim) 🌟🎨🦋🌊🦌

[-] medusa@sh.itjust.works 3 points 11 months ago

I don't even have words, Thelsim, this is awesome 😆 I appreciate the hell out of you, my dude, for putting this much effort into my ridiculous request! Thank you, and ChatGPT, for making my day with this!

[-] thelsim@sh.itjust.works 3 points 11 months ago

It's been my pleasure! Really :)

Stuff like this is always fun to do, especially when I get such a positive response in return!

[-] PersnickityPenguin@lemm.ee 2 points 11 months ago

Wait what? Could somebody explain to me what this means...?

Are you saying that chat GPT can generate images in being, as well as explain to you what it made?

I'm so confused

[-] thelsim@sh.itjust.works 2 points 11 months ago

I guess I can understand the confusion :)

Since the original picture was made with Bing (which uses DALL-E to generate the images), I thought I might be able to reconstruct the prompt by describing elements of the image to Chat-GPT (which also uses DALL-E).
I described the original sketch and the description that came with it, then I kept adding stuff. I told it that the final image is in color, that the grove is set during dusk, that the creature is relaxing in the pond, holding a glowing ball, etc. And every time it would alter the prompt to fit all those elements.
Finally I wanted it to send that prompt to DALL-E and generate the image. Which is where it all went wrong. Ever since an "upgrade" DALL-E seems to be having trouble with generating images. I got very annoyed and tried simpler and simpler descriptions, but it just didn't budge. In the end I just made Chat-GPT write a summary of everything we tried to accomplish and a little apology for not managing to generate the image. That way, I felt like I've accomplished at least something :)

I'm sure there are more "efficient" ways of doing this, but this is a lot more fun :)

[-] medusa@sh.itjust.works 8 points 11 months ago

I GOT REALLY CLOSE AND I AM SO HAPPY THANK YOU GUYS FOR YOUR ASSISTANCE!!! I still have zero idea what happened to my original prompt at this point - I think I must have cleared my search history after having done it - but these great tips did wonders for me!

Bing prompt: deer-skull-headed leshen with mermaid tail relaxes in druid's grove lake. tiny gold chains festoon the antlers on his deer-skull head. He is admiring a glowing orb in his bony hand. magical, serene, peaceful; digital art

[-] medusa@sh.itjust.works 2 points 11 months ago

[honestly the shape of this one is even better than the one I originally shared i love this community so much]

[-] medusa@sh.itjust.works 6 points 11 months ago

This was my original sketch from a dream I'd had. A deer-skull headed skeletal merman with fish tail, antlers festooned with tiny gold chains, admiring golden chains or a gem or something along those lines, relaxing peacefully in the lake of a druid's grove. Very peaceful, very calm, very welcoming overall feeling.

[-] Even_Adder@lemmy.dbzer0.com 6 points 11 months ago

Your drawings are very nice. If you have the hardware for it, you might want to look into getting Stable Diffusion working with controlnets. It'll give you a lot more control over your outputs.

https://www.youtube.com/watch?v=dMkiOex_cKU

[-] medusa@sh.itjust.works 5 points 11 months ago

I've never even heard of this! I'll have to look into it. Thank you for the suggestion!!!

[-] glimse@lemmy.world 5 points 11 months ago

Bing tip: try searching your browser history for words you think you used. The prompt might not be in your Recent Creations list but you should be able to find it that way! I've done that multiple times

[-] medusa@sh.itjust.works 4 points 11 months ago

That's a great idea! Thank you!

[-] Halcyon@discuss.tchncs.de 2 points 11 months ago

In the Bing Chat, there is a clock symbol with an arrow. There you can find your complete prompt history.

[-] Halcyon@discuss.tchncs.de 6 points 11 months ago* (last edited 11 months ago)

My attempts of a recreation:

[-] medusa@sh.itjust.works 1 points 11 months ago

That's the closest I've seen yet! What was your prompt if I may ask?

[-] Halcyon@discuss.tchncs.de 2 points 11 months ago* (last edited 11 months ago)

The prompt was:

"create an image of a deer skeleton sitting in a pond and leaning against a white stone with some green moss on it. In the deer's antlers there is a golden glowing string of lights. In the skeleton's hand it holds a golden glowing ball. Around the pond there are some bare trees and it is foggy. In front of the pond there is a blackbird looking at the deer. There is more green moss on the stones around."

I simply described the main aspects of your image to play around a bit. Bing gave me several other pictures that looked quite different - these two were the only ones where the elements were at the right position and the perspective was similar to the original. It the first pic I didn't have the moss in the prompt so I added that with the second generation.

[-] spyd3r@sh.itjust.works 4 points 11 months ago

[chugs potions and reaches for Witcher sword]

[-] medusa@sh.itjust.works 3 points 11 months ago

water spritz No! spritz spritz Bad witcher! This is an intelligent, benevolent creature!

[-] spyd3r@sh.itjust.works 3 points 11 months ago

How about a round of Gwent then?

[-] FreshLight@sh.itjust.works 4 points 11 months ago* (last edited 11 months ago)

Mine turned out a bit darker..

[-] medusa@sh.itjust.works 1 points 11 months ago

Wow I love these! May I ask your prompt?

[-] FreshLight@sh.itjust.works 1 points 11 months ago

"deer skeleton god mermaid in dark autumn forest lake with glowing orb in hand next to bird" I added "mermaid" to the prompt in the last picture

this post was submitted on 16 Nov 2023
51 points (93.2% liked)

AI Generated Images

7121 readers
64 users here now

Community for AI image generation. Any models are allowed. Creativity is valuable! It is recommended to post the model used for reference, but not a rule.

No explicit violence, gore, or nudity.

This is not a NSFW community although exceptions are sometimes made. Any NSFW posts must be marked as NSFW and may be removed at any moderator's discretion. Any suggestive imagery may be removed at any time.

Refer to https://lemmynsfw.com/ for any NSFW imagery.

No misconduct: Harassment, Abuse or assault, Bullying, Illegal activity, Discrimination, Racism, Trolling, Bigotry.

AI Generated Videos are allowed under the same rules. Photosensitivity warning required for any flashing videos.

To embed images type:

“![](put image url in here)”

Follow all sh.itjust.works rules.


Community Challenge Past Entries

Related communities:

founded 1 year ago
MODERATORS