Create Stunning Portraits With Playground AI | Part 1

[@Ibrahim_Sahil]

👍

[@playgroundai]

🙌

[@POSHREF]

Great video Bro!

[@playgroundai]

Appreciate it!

[@fundacionpersonasqueaman-p7585]

el audio debe tener opciones de idiomas para sacar el 100% el buen uso de las herramientas.

[@Veronikafjord]

I love your voice sooo much!!!!!!

[@playgroundai]

aaawww thank you, you make me blush ☺

[@nks-nqn-xs4yd]

muchas gracias por el tutorial

[@playgroundai]

You're welcome!

[@reyanart3177]

You are amazing!

[@playgroundai]

Wow, thank you!

[@timedriverable]

I got some great results from this but some times I get that "safety issue detected" this is when I went to the analog filter vs the polaroid filter.

[@playgroundai]

If you feel like they are false negatives, feel free to send us the prompt at [e-mail]. We just improved our safety filter and we are still tweaking it. Any feedback is welcome.

[@Cyber_StreeLogs_TechSupport]

Was trying to personify some apps, eg the latest one-THREADS! Wanted to make it as a person, head as threads logo, maybe trapped in threads/ a web, but separate icons not available to personify or animate

[@Jessi-si2kw]

I would never have thought of typing "clear eyes" so thank you for that tip :) I love the AI. Makes very good looking females and males and armor or leather attires. But the hair.. most look the same around the face for both male and female. It's a bit boring. How do I make the AI understand there are different types of hairdos even if the hair is lose and flowing?

[@playgroundai]

you can state the hair style but certain models (filters) will follow it. That has a lot to do with data set they were trained on. Or you can use the base SDXL or Playground v2, v2.5. It will take some experimentation.

[@Jessi-si2kw]

@@playgroundai Hmm. I have to learn the names of hairdos in English. I don't even know most of them by name in my native tongue lol Thank you for the answer, will learn by trying. 🙂

[@sajinscartoonvine]

Sir tell us how can we edit our own photos, like midjourney portrait ............

[@denziioo]

Somebody please explain what samplers are? Nobody on youtube has explained them

[@playgroundai]

Definitely on the deck to do soon!

[@TaviojokerFixas]

Eu já tenho isso num de 3 😅 que abelidade acabada tá como morto

[@patsimon6136]

the best AI that I've seen lately..!!! Bravo Guys !!!!!!and you are the best of the best !!

[@playgroundai]

Thanks so much! The team is working hard on making this a great product. We appreciate the feedback and support!

[@akekitty07]

I want to use the AI filter on me, but it’s not working as well as I thought it would. Can you help please

[@patricial3494]

Just started with this software - unbelievable. I'm so impressed and having so much fun creating characters from my books! Very very cool stuff.

[@playgroundai]

Glad to hear it! Enjoy!

[@sp3lunks]

How do I make a portrait of myself?

[@desmosedici1272]

Hi, I would like to ask you something: can the images that you can obtain with Playground be printed on canvas to make a picture to hang, or on a banner, or printed on a t-shirt? I saw that the image resolution is 120DPI, is there a way to get to 300DPI? with the PRO plan?

[@rebelcommander7starwarsjur922]

Can someone please help me? Literally been trying for hours on end here what I want. A kemonomimi guy that I can use for a roleplaying oc that has full body and tail or at least good face. Here what I get one wolf ear a human bc it has no ears and tail and never once been able to get the tail at all looks weird or is a girl even when my props say guy, male, boy AT THE SAME TIME I did that and like 5 times in a row I got a girl. Here is what I have tried multiple different prompts like kemonomimi, male, boy, guy, wolf ears and tail, wolf ears, wolf tail, full body, person wearing a wolf tail, tail just not all of them at the same time. I have had to also at time negative prompt the thing with werewolf, wolf head, ugly. I have tired almost all filters and changes image strength or quality of image or prompt guidance or model used all but DALL*E bc I cannot pay for shit. WHAT AM I DOING WRONG I NEED HELP! it has clearly worked for others searching kemonomimi guy on the front page tells me that. I need help someone please help me. I just need help I am new to this and I really need help I wanna use this thing but I cannot pay for anything so I hope I do not need to pay but can still do everything I want to do. oOh also I seem to have ran into messages like some error saying I needed to jest perimeters as the AI content system as flagged it or stopped something CONSTENTLY as well.

[@Rherman1950]

Always have a tough time with prompts for portraits. Thanks for giving insight into how to do them better.

[@playgroundai]

Glad it was helpful! Part 2 to come soon!

[@jisumali666]

这也太神奇了吧，我刚接触这个AI，就被迷住，研究了几天，终于做出了个像样的人形图。希望多点分享教学视频。

[@playgroundai]

Thank you, glad you like it!

[@armjee1976]

😮I am initially leaner and watched this video very useful me as basic learn. great effort your team Thanks to all. hope its continue for me this type of learner in the world.Thanks again

[@playgroundai]

You're welcome, glad it was helpful!

[@NanaBampoe]

MIND BLOWN officially !!!!!!!

[@starflower..]

is there any way to keep faces consistent across various generations? i love one portrait i generated, but none of my other prompts really capture the same face in a new way.

[@playgroundai]

Give your character 1-2 names, ethnicity helps too. I have a video to edit on this very topic. 👍

[@starflower..]

@@playgroundai ok, i'll try this, thank you!!!

[@brokenfingers9607]

What a time to be alive.

[@vishalpisote1306]

I am often facing an issue while generating human faces. Most of the time it generates multiple faces. Noticed this several times. What may be causing this? My used this prompt recently - "A beautiful Indian woman wearing Indian Jewellery, 8k, high details, super realistic". I know there may be something wrong at my side at not because of the AI. Already mentioned in other comments that this is the best user friendly AI as compared to others. Please help me on this

[@playgroundai]

So not to get all technical but if you are using stable diffusion 1.5 it's best to keep your dimensions around 512x768 or 640x832 (vice versa). Anything bigger is prone to deformities, it's just the way the data was trained. It's native dimension is 1:1 at 512x512 so the more you sway from those dimensions, the more you run the risk of those deformities. Negative prompts can help minimize it but it can still happen. However with the latest model SDXL you can use bigger sizes like 768x1024 or 1024x1024 as it is trained for bigger dimensions. Hope that helps!

[@GamersTherapy]

FYI - the link to negative prompt list isn't working. Not a major issue as we can still find these, but thought I'd let you know.

[@playgroundai]

Thanks for the heads up Ryan, I've added it to the description.

[@RominaOvelar-g5e]

Muchas gracias, me encanta su página de Playground es mi favorita de todas.

[@playgroundai]

You're welcome! Glad you are enjoying Playground! 👍

[@aigen-learner5356]

I have followed your step-by-step prompts, including the negative ones, but I consistently end up with abnormal changes in the eyes (pupils) or nose in images. By the way, I really love your teaching style! Thank You👍

[@playgroundai]

Thanks for the kind words. What image size are you using?

[@aigen-learner5356]

@@playgroundai 512 x 768 (Sorry for the late response).

[@bengsynthmusic]

@@aigen-learner5356 use the largest size

[@mina_en_suiza]

The video is great, but despite improved images thanks to it, I am still totally disappointed with the results I am getting. Prompts (like smoking, people in background, etc.) are often completely ignored and faces and bodies look mostly horrible.

[@bengsynthmusic]

@@mina_en_suiza Give it an image for it to use as a basis. Increase the prompt following handle and quality on the right side. Add the word realistic in your prompt. Experiment with different style like Instaport for faces.

[@anniegerona2061]

are you using free or pro?

[@timedriverable]

Are you only being charged for the original prompt or every time you create or edit your original image?

[@playgroundai]

Each time you generate an image it's counted, for the edit features some are limited to 30/mth on the free plan, pro/turbo is unlimited at the moment. playgroundai.com/pricing

[@GamersTherapy]

Two tuts in one! How to craft your prompts on playground plus portraits. Nice! Thanks so much. 🙂

[@playgroundai]

Glad it was helpful and you’re welcome!

[@RAFIQMOHAMMAD]

Excellent turorial on the subject

[@playgroundai]

Glad to hear it! More vids on this topic to come soon!

[@speakbot5956]

As already mentioned by other comments, your teaching style is fantastic. I would love to see a lot more in your channel. Topics like photo restorations, special effects, and basically a lot of real world applications for working with client images. One of the most difficult things I'm finding is getting rid of disfigured distortions and other unexpected anomalies using negative prompts. A tutorial on the specifics of negative prompts would be amazing. Thanks again for taking the time to offer you style of training. Much appreciated.

[@playgroundai]

Thank you for the positive feedback and suggestions! We will be posting more content soon!

[@MTB_Nephi]

Wow, thx for it. Fantastic results. I would be interested in prompts that generate certain facial expressions of people. For example: Uneasy look, critical look, joy, sadness, etc.

[@JohnDoe-tx8lq]

I'm really enjoying this app, BUT, I don't see why certain Negative Prompt are not in the coding by default, unless they are specified in thePositive Prompts, or even part of a filter style: Like: deformed, disfigured, mutilated hands, extra limbs, floating limbs, extra fingers, fused fingers, too many fingers... if you wanted those, you'd be actively asking for them!! 🤨

[@playgroundai]

Totally hear you! So firstly, many people create different kinds of images so certain negative prompts won't apply. Also we are working on something that will help with this but I can't say at the moment. 😊

[@JohnDoe-tx8lq]

​@@playgroundai certain negative prompts won't apply - Yer, obviously, like I said.- "unless they are specified in thePositive Prompts," but fused fingers, too many fingers is something someone would rarely ask for, while 10000x more often, they won't want it. That means 10000x people have to specify no fused fingers or not too many fingers for everyone that would want them, who would specify them.. I presume we DON'T have to specify "I DON'T want skin to look like custard" or "I DON'T want the sky to be green" ... seems as likely as wanting fused fingers. 🤔🖐 But it's still all extremely impressive! 🙂👍

Excellent tips! And excellent soothing voice of the narrator! It explains and calms you down! 😂 Loved it!

[@playgroundai]

Glad you enjoyed the tips! And thanks for the compliment 😊👍🏼

[@DandelionStrawart]

These tutorials are very helpful, thanks.

[@playgroundai]

You're very welcome!

[@aegisgfx]

Love you guys, never stop what you are doing

[@playgroundai]

We appreciate your support!

[@david.alexandre.]

Thanks a lot for helping. I did awesome portrait.

[@playgroundai]

Great to hear!

[@playgroundai]

You can upscale the image to get a bit more resolution. It’s the nature of Ai image generation. Most people will use third party upscaling to bring the resolution up.

[@david.alexandre.]

@@playgroundai I've been working in Closeup Shots, and I can already see a very significant improvement with "Portrait Photo". I didn't realize that they are two different things. And about the background light I have also been using "Volumetric light". Great results here. Thanks again

[@brave-MAESTRO]

I want an android app of Playground AI. Currently I'm falling in love to WOMBO AI app.

[@playgroundai]

Thanks for the feedback! We still have a lot of work to add to the site but we'll definitely add it to the list of suggestions.

[@NFTartworks]

I was about to make the same request. In any case it works great on phones but the app would make it a bit more user friendly.

[@miriammiriam8059]

Such a good explenation on how to build up. Thank you!

[@wheinemail]

Thank you for making the Important Playground subtleties so much easier to understand. It would very helpful to me and most likely to others, if you could create a video on merging elements to coexist. I have an assignment to have a birthday party girl with a smaller size cake on top of her head. Playground is smushing the cake down and more behind her head so don’t see the cake but candles coming out of her head. I’ve toyed with bringing a photo of a small cake into playground to see if that might help, but not sure how to merge the two After hours of messing around, I resorted to the cake placement happening in Photoshop. A video of how to prompt extraneous elements around a subject would be extremely helpful. Another example was playing cards flying over a boy’s head. Creating the boy and the girl in Playground was fairly straightforward and they look great. But extraneous elements don’t look even close to what I was attempting. Is that a subject that you would consider?

[@playgroundai]

Absolutely! I've been thinking of a similar video to do on what you talk about. Stay tuned!

[@wheinemail]

@@playgroundai would really appreciate that!

[@MR-ANONYM420]

what about the license? Do you have to buy the pro plan to use the pictures in business?

[@playgroundai]

You can use any image you make commercially

[@adictiv937]

How can I replace the face of a comic character with my own face?"

[@apoemaday]

I am constantly amazed by the new possibilities that gives to the creators. Can you perhaps make a video to show us given sceneries and the method you would use to ''zoom out'' for example, or see the very same elements but from a different angle or different light etc..? (I can't believe I am actually asking this as if it will be the new normal. ) You guys rock!!

[@playgroundai]

Appreciate the feedback and support! We’ll be making more content soon!

[@vladi_litvin]

The tutorial is super. Thank you very much! I would like to see videos about Digital artwork, Digital painting and so on.

[@playgroundai]

Coming soon! Appreciate the feedback!

[@williamflaherty3168]

Waiting to see a tutorial on how to create an artistic portrait with a specific photo as the base. Thanks!

[@playgroundai]

Definitely on the list of topics to cover. 👍🏼 appreciate the feedback.

[@fahzeezstudioz1886]

The face restoration option is good. But, the problem is that option enhance whole the image and make the other things look like painting... The cloths and skin lost its structure and details in face restoration option. Please fix it.

[@djr3386]

Good teaching. Thank you. 🙏 Can we create backgrounds for our photos??

[@playgroundai]

You're welcome! Yes you can just use image to image to mask the areas you want to change, put it in the prompt and run a few to see what you like.

[@hiransoysa7827]

can we develop logos by playground?

[@SnoopyDoofie]

Is it possible to reuse the same character but adjust things like the direction they are facing, the tilt angle of their head and facial expressions? This would be great for storyboarding where you want to tell a story but need the same character.

[@playgroundai]

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.

[@velvetbees]

Thank you for another great video! I hope you channel grows and your website grows too.

[@williamflaherty3168]

Great tips, thank you!

[@playgroundai]

You’re welcome! 🙌🏼

[@ВладСвечников-м9ц]

Hello, could you please tell me how to make an image caption?🙏

[@playgroundai]

Text in AI image generation isn't really possible at the moment. That would have to be done in another program. Ai will translate text but it won't make sense. It can happen randomly though.

[@vaderfan2055]

How can I use photos of myself or family members to generate cool art??

[@playgroundai]

You can use image to image and mask areas of the image you want to change.

[@frymaxa1702]

the API is pretty tight. I tried simple words, not nfsw, but it keeps rejecting the prompts. Anyway, I must say this AI is one of the best

[@playgroundai]

We're always working on improving the API errors as much as we can however it's not directly something we can control completely. Appreciate your patience.

[@frymaxa1702]

@@playgroundai nevermind. Your platform is one of the best!

[@sigmabrindes]

Obvious that make a Photo is very important, but I'd like to suggest make some art composition with Cartoon (like PAW Patrol, PJ Masks, Princess Sofia, etc.) in a scene like an "art" to put it in the MUG or something like that!

[@jinjiang3875]

The face of the character will still appear deformed when using these vocabulary words in a full-body portrait. A half-body portrait might be better. What should I do?

[@anniechuachannel]

Hi, I've been trying many times in canvas to erase the necklace on the dog's neck and have given prompts as well, but it still doesn't work, it just changed the style of the necklace😅. May I know what I should do?

[@playgroundai]

Hi Annie, there is a few ways to approach this. You can try inpainting and mask the area but remove the prompt or at least simplify it. If that doesn't work use the edit function to remove it. We are working on better and easier ways to do this!

[@anniechuachannel]

@@playgroundai Ok, will try, thanks.

[@ahmadraza8137]

How to use the same character for multiple portraits?

[@playgroundai]

We will cover this in future videos!

[@Marksplaytime]

Probably batting out of my league, but the issue I'm having is trying to remaster and perfect an existing picture that was maybe really grainy, poor lighting, scanned photograph, or even a screenshot from a digitized home movie where even Photoshop is only getting so far. Making new random people out of thin air is one thing, but remastering then reimagining an old picture of my wife would be far more rewarding (and probably lead to fewer questions).

[@playgroundai]

Marksplaytime, this is something I've been experimenting with personally. It can be done to some extent but would still require a bit of post production. I've been doing my own testing and plan to create a video on it in the near future. As the technology advances, it will only get better and easier to do. Appreciate the feedback.

[@Marksplaytime]

@@playgroundai I'm starting to have some success between the Playground and Photoshop and have produced a couple things I'm really happy with, but I have to start with a pretty good picture. Otherwise, yes, I'll be adding this to the list of things waiting on better tech. I look forward to seeing what you come up with.

[@ryanhowell4492]

Cool Tools

[@texasenglishacademy5302]

I love playgroundAI, it free and unlimited generations. I use it to generate visuals for my esl students. I tend to cater to university students and working adults and try to keep my cost low (I'm poor). What I would like to see is the option to save settings. Think about it playground is generously using a lot of gpu power with people like me constantly generating images until we get it right. Then we log off later come back and start the process all over again, generating images until we get it right. Save settings for log in users would be a wise idea. Thanks for the video.

[@playgroundai]

Great feedback, I'll definitely pass it off to the team. The settings are saved per session but they do reset once you leave the site.

[@pluckypluckster]

oh that's disappointing i didn't realize that oil painting gave that exact style - thought it was something i randomly did. its not really even oil painting looks more large stroke acrylic to me. When i think of oil painitng i think of smooth transitions, small unnoticable brush strokes. :( good video though!

[@playgroundai]

You can add more to the prompt to alter the style a bit more.

[@faridagadiri1970]

C'est fantastique

[@JimArthurVanWyck]

Wow…another terrific tutorial. If I may, I would love to see a tut on placing a product phot in different backgrounds. Very useful for product reviews. Anyways thank again. I subscribe to this channel and your personal one too.

[@playgroundai]

Hey Jim, always great to hear from you and thank you for supporting! Great suggestion for a video topic. I'll add it to the growing list. 👍

[@olivierduveau1518]

Bonsoir, comment créer deux personnages ou plus sur la même photo ?

[@ohioinsuranceDOThealth]

Super impressed!

[@playgroundai]

Thank you for watching!

[@prazdnichnyy]

I'm falling in love to WOMBO AI app.

[@pawnoir]

Merci👍

[@TheCreativiZt]

Why don't you guys default those exclusions?! Who would want extra body parts and a women in the shape of a figure eight pretzel? Allow members to unlock them by adding instead of forcing us to exclude them. Edit: and 50% of my photo / portrait prompts keep coming back with "safety issues". So annoying. I just want a woman in a dress. Not nude. Not with anything that would trigger a "safety thing". But the absolutely most annoying thing about Playground AI is all images with people create duplicates. I just want solo people and end up getting 2-4 of everything. It's the only thing that prevents me from promoting this site on my other youtube page. Edit: By the way, I think someone is sabotaging your site. Type in "beautiful" or beautiful woman" and you get some black mans face with skin warts and no teeth. Type in "woman in bikini on beach" and you get an 80 year old woman with fat and cellulite. You might want to address that. I think people are setting Playground AI up for failure.

[@playgroundai]

We are always looking at ways to improve PG. One of the main reasons we don't put negatives by default is because there are many subjects people generate. Personally I keep a document of my most used negative prompts and copy and paste but I like the idea of being able to "save" them somewhere to use with a click of a button perhaps?

[@TheCreativiZt]

@@playgroundai Yesssss please. The ability to save our entire create page layout as a draft would be a God sent.

[@MrX-nh1xm]

0:02 , that female Deadpool made by me :))) btw i want to ask, once we restore a face, we have to download it , i usually do that once im happy with the final result, if i bring it back to image to image , again with the default image strength of 30 it shows a diff variation. i want to keep the same restored version with the restored face and sharpness, for now i increase the image strength to retain it, i don't know if im missing something here , is there a way we can keep the restored or upscaled final version inbuilt changes, im new to playground

[@playgroundai]

That's a great image! Nicely done! As for the face restoration, at this time the workaround you mention is required should you want to reuse the image for further creating. We know that this is often requested.

[@ВладСвечников-м9ц]

Hi, I saw your work in playground Ai, very amazing. Please tell me how do you leave your signature ”MR X" on the images?

[@MrX-nh1xm]

@@ВладСвечников-м9ц thanks, i actually made a custom signature png in photoshop,i just use that to put on the image :) , i think bella and vinchi, and ppublish use the same way , because there is no option for text in playground, unless they are using the paid version and it has that option

[@ВладСвечников-м9ц]

@@MrX-nh1xm Thanks for the answer, I realized that there is no option to sign the image on the playground. I'm interested in something else. In order for the image with the signature to be on the playgraund, do you download it, insert the signature and upload it back to the playgraund or how to do it?

[@MrX-nh1xm]

@@ВладСвечников-м9ц yes, when i am happy with a image , i do face restoration or upscale as final steps, then i will download it as it gives that option, i thn put the signature in photoshop , save it there and bring the image back to playground

[@eprahs1]

PlaygroundAi, please please get rid of the endless scrolling feature on the profile page. Please do multiple pages instead on endless scrolling

[@playgroundai]

Appreciate the feedback and we are definitely working on it.

[@ArtBuddy-bf1is]

PART 1 PART 2 THE SAME NO ?

[@PosivibesOnly2024]

I have only discovered y'all recently and its very addicting to make different prompts and things really loving the way AI works in photography nowadays at least!! I am excited to see what types of things you shall bring for us to play with in the future!!! sub'd n lov'd!!! ♥ p,s thanks for the negative prompt i always have problems with weird things in my pics!! :)

[@intecturentv2907]

Excelente Playground AI... enhorabuena!!! Saludos desde la destruída #Venezuela por culpa del #Socialismo #CastroChavista, y q esperemos salga este gobierno de facto en las próximas elecciones!

[@claudianreyn4529]

It's not an interest in making portraits, it's a limitation. Stable Difusion makes very bad faces with defects if they are not big on the canvas.

[@playgroundai]

There is a limitation with Ai generated images in that sense. However there are work arounds in post production that can be done.

[@NFTartworks]

Sick and tired of these portraits. As if people don't have any imagination anymore.

[@playgroundai]

What would you like to see?

[@NFTartworks]

@@playgroundai countless other options. It's just the thing with the portraits that's bothering me like when everyone was creating crypto Punks and Bored Apes copies a few months ago (some still do). Just show the world that Playground can create anything!