2024-10-01 Synthetic Photography SIG meeting notes

We met online using ZOOM 6:00pm-8:30pm

You are invited to join us on ZOOM for the "Synthetic Photography SIG". I expect that this will be a controversial SIG as the topic of AI generated art seems to polarize people into 2 camps, they either love the concept that now they can make beautiful art or they hate the idea that "unskilled" people can make better art than they can. I am sure that we will have different views, but I hope that we can get along and respect everyone's views.

"I Disapprove of What You Say, But I Will Defend to the Death Your Right to Say It" - Evelyn Beatrice Hall (not Voltaire)

ATTENDEES
Mike Barry
Mark Bolles
Jim Fellion
Jim Limburg
Jack Lipscomb
Walter Lyons
Paul Marcoux
Greg Nott
Ken Sandberg
Albert Wang

1. We discussed FLUX
We discussed image created by FLUX, and I think that everyone agreed they were better than most of the images created by most of the other text-to-image generation programs. Walt showed quite a few very creative images and we discussed whether some of the images would pass for regular photography. FLUX quit working so we were unable to do any live demos.

2. We tried out the FLUX Prompt Generator
It does make the prompt more detailed, but sometimes it adds details that weren't asked for. Overall it could be a good tool to stimulate your creativity. Since FLUX was not working, we used the generated prompt in DALLE (Microsoft Copilot), and it did work but the image quality was not as good as FLUX images.

3. We used Leonardo to demonstrate img2img, or image to image generation and it was quite good.

4. We looked at a couple new tools:
Freepik - free online interface to the FLUX tool
Hotshot - free txt2vid tool - amazing, but not quite ready for prime time
Minimax - free Chinese txt2vid tool - amazing, but not quite ready for prime time
Adobe Firefly announced that they will be releasing a txt2vid tool in the next couple months

5. We decided to create composite images using AI generated components for the next meeting. You can mix photographic components with AI generated components, or just use AI generated components. I plan to composite them using Photoshop, but you are welcome to use any tool maybe even one of the AI tools. One of my goals for AI image generations is to generate backgrounds to put model images in.

RESOURCES (Where you can find sources of inspiration)

Minimax - txt2speech and txt2vid - Chinese based
https://www.minimaxi.com/

Hotshot - txt2vid
https://hotshot.co/

FreePik - FLUX based txt2img
https://www.freepik.com/

Prompt suggestions
https://openart.ai/promptbook

Sample prompt styles
https://supagruen.github.io/StableDiffusion-CheatSheet

Stable Diffusion XL Artists List
https://stablediffusion.fr/artists

FLUX online free version
https://FluxPro.art

FLUX Prompt Generator
https://huggingface.co/spaces/gokaygokay/FLUX-Prompt-Generator

AGENDA FOR OUR NEXT SYNTHETIC PHOTOGRAPHY SIG MEETING - Tue 11/5/2024 @6:00pm
1. Discuss composite images that we created with AI components
2. Demo A1111 with stable diffusion running on my local computer
3. Look at CivitAI for prompt ideas
4. Discuss questions, ethics, techniques, what is happening with AI in general

Please email Mike, info@fcdcc.com, when you find mistakes, missing information or if you have suggestions for the Synthetic Photography SIG and I will try to address the issues.