Get Creative With Image to Image & Inpainting in Playground AI
Вставка
- Опубліковано 24 лип 2024
- Get Creative With Image to Image & Inpainting in Playground AI. In this video we will get you started on using image to image and inpainting to further enhance your image generating workflow.
Playground AI is a web platform based on stable diffusion to create text to image digital artwork. playgroundai.com
Prompt: (playtoon filter) cute and adorable racoon wearing a suit and top hat, anthropomorphic, standing on the sidewalk
Negative Prompt: text, signature, title, heading, watermark, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, out of frame, ugly, extra limbs, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, mutated hands, fused fingers, too many fingers, long neck
0:00 Settings and a simple prompt
2:20 Variations
3:00 Image strength examples
3:50 Inpainting/Mask
5:40 Draw to image
7:53 Prompt the drawing
Find us on
Playground AI - ai image generator: playground.com
X (Formerly Twitter) https: // playground_ai
very grateful for this video! It really helps in fine-tuning a specific image withour a time-consuming loop of every-newly-made images. Thank you!
I call this my daily AI' course'. Love it, especially finally being able to create the images that I can never get on paper with a pencil! Thanks so much for your amazing videos! Great voice and no mind numbing tunes!!
Great to hear and yes that's the whole point of AI image generation, to be able to create the ideas in your head and make them reality. Glad you are finding the videos helpful and I'll make sure to keep the mind numbing tunes to myself hahaha!
I want to thank you very much for what you do, playground ai is in my opinion the best AI that exists today. I wish you to conquer more and more heights in the development of playground ai. May generation be with you :)
I recently started messing around with Diffusion AI and your video came up in my feed. So glad that:
A) You took the time & effort to make this video tutorial.
B) That I've watched it start to finish, twice in a row.
Superb presentation at a pace that we can keep up with as well as providing better insight as to how to manipulate the tool & generated images to get closer to one's initial vision/idea.
Needless to say that as a result of all of this, you've garnered yourself a new subscriber. Really looking forward to seeing some more advanced topics in the future (i.e. the differences in the renderers, a brief description of what each filter does including their stengths/weaknessess, etc...)
Awesome. Simply awesome!!!
Thank you for the great feedback! We will start posting more very soon so stay tuned as we have lot's of topics to cover!
I mixed the 2 works and the result is amazing, instead of a racoon I used a squirl
I have made my initial image to image picture in AI first. Then I used it to create other images. I try to make a simple model of what I want. It works really well. Also, the image strength slider under the image is a valuable tool. I slide it to a low number and it gives me a variety of different images to choose from. Imho, Playground is the easiest to use for great results.
Great tutorial, I didi not know can drawing the picture, thank you so much!
Yes it's a fun way to create images as well and you can get some interesting results with it!
Great video. “That’s what she said joke was hilarious!!
hahaha! I totally forgot about that 😆☺️
thank you very much. i love using playground ai, i think it's a very successful app. it will be great if it can create the same characters in different environments.
This can already be done with utilizing seeds however with any ai generator at the moment, to get 100% consistency in regards to clothing, attire even character styles is not exactly possible. You can get close but not always consistent. With that being said there are work arounds and we will cover this on the channel very soon. The technology is advancing very quickly so I can only assume we are not far off from making this possible.
Playground AI is amazing i lke it most than Midjourney
I like your videos,you are a real teacher,thankyou for your patience and the way you explain 🙂
Very kinds words and thank you! Means a lot to me. Stay tuned as we will have more to come and on a more consistent basis!
👋Welcome!! and yes i will still tuned to learn more 🙂👍
I haven't watched the whole video yet, but I'll hit like.
I haven't read your whole message but I hit like regardless 😁
@@playgroundai Now I watched and .... Great! Thanks for sharing
hice el mapache y el paisaje pero lo de la mascara no me sale , borro todo pero no me crea otro escenario, no sé donde estoy fallando,😅
Cool tools
Love your tutorials! Question: when I want to further edit the last painting (end of the video) I'm happy with ....... add more details or correct things on the upscaled painting, what do I need to do?
Thank you! If you are on desktop/laptop it’s best to bring it into canvas, there you can add details, expand remove objects etc. if possible I would only upscale the final image last when there are no edits to be made. Hope that helps!
Wow, thank you!
You’re welcome! 👍🏼
In Canvas, how can I connect 2 similar images together, especially the pattern on each image.
I want to join them to be one big endless flow pattern.
What prompt should I use for inPaint generation frame?
do u know you are the best of the best AI in this direction ??? 😉😉😉
We appreciate the kind words and support Pat. More to come soon! 🙌
Nice 👍
Thank you! Cheers!
Thanks guys this is awesome
You are welcome!
Hi, there isn't "Done" button. may be it was changed?
Using the mask feature,s is it possible to use PlaygroundAI to adapt a photo of yourself to add a Parisian cafe setting for example?
Technically yes, you would just inpaint the background and put the cafe in your prompt. 👍
💙Great video. Thanks! Great videos, Quick question: Bellow Image to Image there are 2 options which seem to do the same thing: "Add Inpainting Mask" and "Edit with Mask".
What is the difference? Thanks so much!
Great question...so the inpainting mask is most common and will give you better details with your prompts. This model will be updated soon btw. The edit with mask is more for simple quick edits like color, small details etc. Prompting for it is simple as well. However not as effective as inpainting. This video explains it, it's slightly outdated though ua-cam.com/video/JFt_YFWJu8o/v-deo.htmlsi=msgsmrwF9twX5vUO
@@playgroundai awww thank you!!!!!!
fabulous to see this process - i dont see option to choose which model - has something changed since this video and canvas is fixed to a certain model
Wow you've been busy watching our videos! Awesome! So this video is on board, it's a separate interface for quick generations. You can select the model on board whereas on canvas, it chooses automatically depending on the filter you use. If none is selected it defaults to SDXL 1.0 which is the latest model.
@@playgroundai really appreciate your feedback! You mentioined a private mode and how that affects NSFW but I see no such mode - is that because i would need to become a paying member and subscribe or am i missing something> thanks again!
If you are on Canvas, all images are private until you manually publish them. On board, there is a toggle on the bottom right. Sorry should've clarified that. 👍
@@playgroundai amazing thanks so much im loving these tutorials and insights your feedback is invaluable -
Use same character in different images to make a story
how to create a human figure seperately and background seperately and the merge together to form one complete image?
You can try this method but I'm also working on another video using img2img instead of outpainting. Coming soon! ua-cam.com/video/bV2E7CAd2u8/v-deo.htmlsi=tBMTClX96pp1HkSb
I’m loving playground, it’s easy to use for a noob like me. I’m struggling to get a hand drawn sketch type of look. Any suggestions?🇦🇺❤️
Great to hear! You can try those words like sketch, pencil drawing, line art and if you have an artist in mind you can use their names in combination. Also play around with the samples too as they can have a bit of a different affect. It may take some experimentation.
@@playgroundai cool, will give line art a try. Thanks for the tips 🇦🇺❤️
what a lovely man ☺
☺😁
I really like this but the issue I have is, when I use inpaint the image quality decreases. Is that supposed to happen? The crispness goes away and it’s not as good as the original. What am I doing wrong?
Make sure you are using inpaint and not “edit with image”
@@playgroundai Maybe I wasn't paying enough attention to the video, but I'm still not really clear on the difference between inpaint and "edit with image". Is inpaint used more for adding things, and "edit with image" more for removing/altering things?
Is there a way to inpaint using a prompt, is it possible to inpaint using an image? For example, including a character or product into the scene of a generated image?
Hi Hajar, yes however it’s best to do this in Canvas as it has more tools and flexibility. I’ll be covering this on my next canvas video. 👍🏼
@@playgroundai ah, so post-processing instead of being part of the AI generation process? Will look out for it. Thank you
How do I use text in my image's as it's creating text in AI language. Thanks
I just did a video here ua-cam.com/video/yWa7BlEpUiM/v-deo.htmlsi=qevkDfUoRcU-V9Cv
Home page should only show images generated using only prompts and originally generated images. Too many images don't have an OG image link for remixing and inspiration. Feels like false advertising if I'm being honest.
Appreciate the feedback. Users are free to do as they please with their prompts however we are hoping these tutorial videos will help people to understand how to prompt and utilize the tools for what they want to create.
is it possible to bring a charakter into a new scene? like from a walking man to a sitting man? i want to tell a longer story with the images.
Yes there are workarounds for this however getting consistent clothing or a specific look is a bit tricky since AI images are random by nature. There are various ways to approach this such as using seeds, or for consistent faces you can use a made up name. I will cover this in a video to come soon!
@@playgroundai nice im looking forward to it
Can i use this image for my youtube videos or any commercial use?
Yes you own the rights, please check out TOS help.playgroundai.com/en/articles/6848773-how-does-copyright-work-do-i-own-the-images-i-create
@@playgroundai Thank You,
Actually I'm an 2D animator, i need some background for my stories. I tried this AI but didn't get good results.
I tried to work with the program for a few minutes and I got this error, I can't do anything else, neither log in nor log out: Error 500 (Internal Server Error)!
Unfortunately we've been experiencing some 500 error spikes today and our team is looking at it. Seems to be sporadic and you should be able to log back in after a few minutes. Please be patient while we get to the bottom of it. Thanks for letting us know!
would like to know how to create animation
thank you for the wonderful site
Currently animation is not possible with playground at the moment.
Ok thank you for your support
we need more
More to come soon my friend! 👍
magnificent video, but how to unlock canvas tutorial
You have to complete the tutorial