14 күн бұрын
3 ай бұрын
3 ай бұрын
5 ай бұрын
5 ай бұрын
5 ай бұрын
Жыл бұрын
Пікірлер
@turgay9956 5 күн бұрын
Hi Sebastian, is it possible to add custom background image and blend it with a subject to create a realistic photo? I don't want to create ai generated background, I need to use my own background.
@ThinkDiffusion 3 күн бұрын
Yes, of course - the background image doesn't have to be AI generated at all. You might need to play around with a few different models to find a realistic one that will generate similar inpainted images to the background. You can also join our Discord and ask for any help here: discord.gg/hAm87ApunD
@BrookeKemp-q7p 6 күн бұрын
Appreciate the detailed breakdown! I need some advice: My OKX wallet holds some USDT, and I have the seed phrase. (alarm fetch churn bridge exercise tape speak race clerk couch crater letter). How can I transfer them to Binance?
@RapidReachAI 13 күн бұрын
The way you integrated the QR code looks super cool might try this out myself
@ThinkDiffusion 7 күн бұрын
Yes, it is super cool! You can join our community on Discord & send what you create here: discord.gg/hAm87ApunD
@JeffreyHarrington 14 күн бұрын
Thank you!
@ThinkDiffusion 14 күн бұрын
Of course, glad you found it helpful! Happy generating:)
@Jacksmith-qu9ii 18 күн бұрын
nice work do you think that I could run this work flow on a linux 12gb Vram setup on a nvidia 3090 Ti ?
@ThinkDiffusion 18 күн бұрын
Hi there! Yes, it may be enough - although will be definitely slow. You can run it without any problems on ThinkDiffusion on this machine: www.thinkdiffusion.com/select-machine/featured/comfy/beta/turbo Happy generating!
@DisgustingJustinAD 22 күн бұрын
Is ThinkDiffusion a locally installed version of Stable Diffusion? Or can I only use it through internet access? I get a lot of heavy storms in my area & lose access to internet during those times.
@arsletirott 28 күн бұрын
fantastic. thanks for the guide. I just wonder why you choose "default negative" aside from "digital painting", i am new to this and ive never used "default negative", what does it do? EDIT: it worked well with the word "cowboy" but when i type "green monster boy" the result becomes a mess and not at all like the cowboy ballerina
@ThinkDiffusion 20 күн бұрын
Hi there! Default negative is a preset of negative prompts that works well for most of the images in sd 1.5. That is why Sebastian chose it :)
@Lobster_With_A_Gun Ай бұрын
I fucking hate ai
@Cu-gp4fy Ай бұрын
Can I get a koooohyaaa
@ThinkDiffusion Ай бұрын
Love it! 🙌
@maziar1382 Ай бұрын
hey more dad jokes, please
@ThinkDiffusion Ай бұрын
🫡🫡
@maziar1382 Ай бұрын
@ThinkDiffusion BTW thanks for those free stable diffusion styles
@darbycarpenter3032 Ай бұрын
I have the Canny button but there is no model. I went to the model folder and nothing is there. All that I have is the openpose. Could you post a link to the Canny model?
@ThinkDiffusion Ай бұрын
Hi there! Of course, here you go: huggingface.co/lllyasviel/sd-controlnet-canny Happy generating!
@paul.tergeist.41 Ай бұрын
A huge thank you for helping the new creators on A1111, Sebastian, because there are really not many complete tutorials. Personally I started in May, almost on my own because I really had to look for information everywhere. A1111 has excellent potential, despite the bugs of the hands etc..... while waiting for SD3. Congratulations and good luck with your channel... 👍🙏🙏🙏
@ThinkDiffusion Ай бұрын
Awesome to hear that, we're glad it was helpful and good luck with generating:) Any specific tutorials you would want to see? 👀
@paul.tergeist.41 Ай бұрын
@@ThinkDiffusion A video, for example, on where to find styles and renderings.... and have a simple and magical and detailed method to add styles (why change the files. CSV and how)..... in short, something simple and useful..... 🙏🍀🤞
@Cu-gp4fy Ай бұрын
Pretty coolseems to be a lot of buttons to play around with and customize
@ThinkDiffusion Ай бұрын
Yes, it's super cool!!
@chrismall-x6g Ай бұрын
fantastic! it was so interesting to listen to. i wanna play it
@ThinkDiffusion Ай бұрын
Thank you! Yess, we can't wait to try it out as well 🙌
@Cu-gp4fy Ай бұрын
the game art looksgreat love the hidden image look pretty trippy
@ThinkDiffusion Ай бұрын
It really does, so creative! 🧙♂
@stableArtAI Ай бұрын
reference to good text, enable T5 for text although SD does know many words which is.a big part in creating text objects.
@stableArtAI Ай бұрын
Our fist run with flux did not yield as good result as there online demo. We might put some time into later or early next year as it continues to evolve. It was though pretty easy to get it running under OS X.
@ThinkDiffusion Ай бұрын
Hi there! If you need some help you can always join our Discord here: discord.gg/hAm87ApunD See you in there:)
@Ai_mayyit 2 ай бұрын
Can we change the outfit with this workflow?
@ThinkDiffusion Ай бұрын
Yes indeed you can change the outfit, you can also change the scenery and the face:)
@gandonius_me 2 ай бұрын
Hello, I would like to ask this question. Why do I set openpose_full and control_v11p_sd15_scribble [d4ba51ff] then I get the skeleton of the image (source) then I click generate, but the image is created according to the prompt without a skeleton, it’s just created according to the prompt
@MartinBenesCreative 2 ай бұрын
Very good video. Straight to the point. Keep going with this channel. It will grow fast! Good luck buddy! 🎉
@ThinkDiffusion 2 ай бұрын
Thank you Martin, we appreciate this a lot especially coming from you! Have a great day:)
@4thObserver 2 ай бұрын
I usually do manual inpainting for fingers instead, Much more accurate that way and gives us humans something left to do. (Lol.) ControlNet I'll use for when I want a real photo as the pose reference or a specific architecture as background because you can layer these together.
@ThinkDiffusion 2 ай бұрын
Good point, yes it can be more accurate and if you enjoy the process of doing it that's all that matters!
@Truthseeker_12638 2 ай бұрын
Hey can you please post the workflow somewhere easy to access I can’t find it
@ThinkDiffusion 2 ай бұрын
Hey there! Yes, you can download the workflow here (check the comment section): www.linkedin.com/feed/update/urn:li:activity:7240010660099616768
@Cu-gp4fy 2 ай бұрын
Super creative application of thisworkflow thanks for sharing guys! could use skme extra cloud help my laptop is garbage
@ThinkDiffusion 2 ай бұрын
Thank you, glad you liked it! You can try out ThinkDiffusion today (free trial) here: bit.ly/3XfyWzt 🌟
@videoaccaunt 2 ай бұрын
deep dive....3:25 min
@ThinkDiffusion 2 ай бұрын
It was almost a year ago 😅 should we record an updated version that is more in depth? :)
@letsgoletsgoletsgoletsgoletsgo 2 ай бұрын
i have a very specific question , i am a photgrapher for a shoe company , i shoot a lot of white background ecommerce photos of the products . ideally i want to input those ecommerce photos into an AI platform , generate a new image with a fantastic background WHILE retaining 100% how my product looks like . is this possible with img2img ?
@ThinkDiffusion 2 ай бұрын
Totally! Check out this tutorial: learn.thinkdiffusion.com/bria-ai-for-background-removal-and-replacement/ we can also use IC-light to relight the subject to match the new background: x.com/thinkdiffusion/status/1806347642550600004 Hope this helped:)
@letsgoletsgoletsgoletsgoletsgo 2 ай бұрын
@@ThinkDiffusion thank you so much, a second question, is there a paid service I can use on line ?
@erutan108 2 ай бұрын
Thank you so much for having this session with Sebastian! More like this please ;)
@ThinkDiffusion 2 ай бұрын
Thanks for the positive feedback! More on the way, very soon 👀
@erutan108 2 ай бұрын
@@ThinkDiffusion thank you! Looking forward to it 👀 and can’t wait to try it on your platform 😁
@Cu-gp4fy 2 ай бұрын
seems to have more control than mid journey nice!
@ThinkDiffusion 2 ай бұрын
Yes, with Stable Diffusion you have way more control than any other AI image generators:)
@Vanced2Dua 2 ай бұрын
Mantap... Lanjutkan tutorial A1111 saya sangat menyukai
@ThinkDiffusion 2 ай бұрын
Thank you for the positive feedback, more tutorials coming every week!
@jmcasler1512 3 ай бұрын
Sweet!
@ThinkDiffusion 2 ай бұрын
😄😄
@whootoo1117 3 ай бұрын
Only black man you have must be ape. What a suppression to love of black men you have. It is reversing it so nobody know your obsession? What is this hate against black muscles, masculinity & power you have? Good luck with it. It is like Hulk of Marvel studios becoming a green figure muscular alien instead of a black man in 20th century. Insecurity of some male races?
@camchanimation 3 ай бұрын
Where can I find the styles you're using?
@ThinkDiffusion 2 ай бұрын
Hi there! They are available for Sebastian Kamph's Patreons:)
@Boxels 3 ай бұрын
I'm hooked! How do I pay you to help me build a model to help me train the AI to make my style so I can animate my stories?
@ThinkDiffusion 3 ай бұрын
Hi there! You can contact us here: www.thinkdiffusion.com/studio Have a great day!
@Boxels 3 ай бұрын
so helpful. keep it coming. how do we get a tutorial on animating for story telling like my channel ? kzbin.infoiZ_UnzQTOVg
@ThinkDiffusion 3 ай бұрын
Awesome to hear you found it helpful! Sadly I am not able to open the link you attached...
@Boxels 3 ай бұрын
exactly what i've been looking for. knowing its possible - but getting lost in tutorials, this is a great help so far!
@pescemagicquo2900 3 ай бұрын
not about the video,but how can i delete my account?
@ThinkDiffusion 3 ай бұрын
Hi there! Please send an email to [email protected] and we'll take care of it!
@HacknSlashPro 3 ай бұрын
so sebastian kamph channel is abandonned now?
@ThinkDiffusion 3 ай бұрын
Not at all!
@jonahoskow7476 3 ай бұрын
wait.. what does an embedding do? Thanks!
@ThinkDiffusion 3 ай бұрын
Hey! Stable diffusion embedding is a method that helps to reduce the complexity of large sets of data by simplifying them into fewer dimensions while keeping the important patterns intact. It’s particularly good at handling noisy data, making sure that similar pieces of data stay close together in the simpler version. This technique is useful for visualizing and identifying clusters or groups in your data. Hope this helps!
@MsMinnyMod 4 ай бұрын
Thanks for the tutorial. I will try to make an anime movie with my kids following your video. We'll see how it goes.
@ThinkDiffusion 3 ай бұрын
That's awesome, keep us updated!
@cfx3 4 ай бұрын
Good video, I feel like the audio track of your voice could have been a bit louder compared to the background music.
@ThinkDiffusion 2 ай бұрын
Thank you! Yes, we'll be working on fixing that for the future videos:)
@ShengzhuPeng 4 ай бұрын
Hi! I’m interested in a business collaboration. Could you please share your email Thanks
@Designing-hc5pz 4 ай бұрын
when will you gonna release new model and it's about sd3 or sdxl?
@matheusreisfernandes8463 4 ай бұрын
When will the When will the pony diffusion be available?
@deefster74 4 ай бұрын
why 1.5 vs XL? seems odd to release a new workflow on an old model
@bgtubber 4 ай бұрын
Audio is too low. I could barely hear anything. :(
@ThinkDiffusion 4 ай бұрын
So sorry about that! It'll be fixed next time we do a workflow video like this:) Hopefully you're still able to hear enough to try it out!
@bgtubber 4 ай бұрын
@@ThinkDiffusion Yes, I had to crank it up and then it's audible. Thanks!
@Pellapoo 4 ай бұрын
He named the workflow "Mein Kamph"
@MultiOmega1911 4 ай бұрын
bruh
@Cu-gp4fy 4 ай бұрын
Thx for the workflow ! are there any other prepicessors u reco? noticed canny but not scribble
@ThinkDiffusion 4 ай бұрын
Scribble is great too, especially with the xdog preprocessor!
@Mel-vc7cs 5 ай бұрын
Hello, I don't see all the sampling methods you are showing in this video. for example, I just see DPM++ 2M not DPM++ 2M Karras. Any idea why? Thanks
@Mel-vc7cs 5 ай бұрын
I don't have the roop option. can you explain why?
@rukakun 5 ай бұрын
sorry if this is a stupid question, I'm a complete newbie. But, how does Think Diffusion differ from using stable diffusion? I'm (getting) familiar with Stable Diffusion, and am struggling to find what I can use ThinkDiffusion for other than generating images...
@ContDeekBruv 5 ай бұрын
This is like the Paralympics