Leonardo AI Image-to-Image Tutorial

[@excelerator]

Leonardo AI has updated many areas of the user interface. See my update video for more information: kzbin.info/www/bejne/eJSliqGNqrKre5Y Use Leonardo AI (FREE) 👉excelerator.tech/GetLeonardo

[@NinjaToddAdventures-wq2zw]

Hey Mr. Excelerator, thanx for the great tips. It is very frustrating getting into AI as everything/the apps, change so damn fast! Thanx again

[@excelerator]

You're quite welcome! And, yes, these things change often!!! I have made tutorial videos and before I can get them edited and uploaded the next day, the platform changes so much I have to start over. But it's kinda fun too. 🤷🏽‍♂️

[@AIdevelopsyou]

thank you.

[@Vicky-yu8rz]

wow! really love the tips! i also want to learn other ai tools? do you have any recommendations? im currently using Stylar and would love to know some of the tips!

[@excelerator]

Thanks. I'll check out Stylar. My other recommendations would depend on what you're looking for: images, video, writing, productivity, etc. Check out my toolbox at www.excelerator.tech/toolbox for a bunch of great tools.

[@Lenpreston2]

How would I colorize my black and white line art cartoons like you demonstrate at 9:40!in your video. I have tried to do this, and my results haven’t been good. Can you help? Thanks

[@excelerator]

It might be too much to type, so I'm making a video as an answer to your question. I'll get it edited and published in the next few days and let you know.

[@excelerator]

Here's the video I made to answer your question: kzbin.info/www/bejne/maqYm315iLqfa7c

[@daliy-kids]

hy sir your video is very informative but my client wants 1 child images that show different grades and ages same face I can't do this please help me and guide how can I do it in Leonardo or any other tool that makes it perfectly

[@excelerator]

There are some platforms/apps that only do this such as wondershare ai lab. I'm told that midjourney will do it if you supply an image and say "image of this person at ____ years old." I have not used it so I cannot confirm if it works. In Leonardo, I suggest generating an image to start with. Then, use the seed from that image to generate images of the older/younger versions. You might also try character reference, or a combo of character reference and fixed seed. In each set of generations there's likely to be one that would pass as a year older or younger. Maybe use that one as the base for the next age. Experimenting with weight values will likely be important too. For instance, the weight of the reference image, the prompt weight, etc. a small adjustment can make a huge difference.

[@dzordzuu5571]

how to change the background, e.g. keeping the character from our uploaded photo, so that the character is almost identical and the background is different?

[@excelerator]

I might be thinking sideways but, if the goal is to end up with the same subject and a different background, wouldn't the easiest route be to use a background remover and then replace the background with whatever you want?

There is no "image guidance" in my account. Are you using a subscription instead of the FREE account?

[@excelerator]

I am, but I thought the free plan allowed it. However, this tutorial may be prior to the redesign of the image generation page. On the left, switch the preset to anything other than Phoenix (that model doesn't support image guidance yet) then look for a faded icon like an image inside the prompt box toward the left. Click it and you should see image guidance options. Or, in the upper right corner of the image generation page, click the toggle that says 'legacy" and then everything should look and work like my video shows.

@@excelerator yeah, thought so. The FREE version is maddening. I need to create a book cover for a word search puzzle book for adults. I was trying to do this in Canva (i'm a total newbie) but it took hours and the result is .. ;meh... Could be much better. So, I started looking towards A.I. and Leonardo looks nice but I have no idea how to start and what to do. Would appreciate some help if you're willing. I'm not selling yet, so I won't be able to invest anything in those softwares like book bolt and others.

[@excelerator]

Ah, I see the dilemma. Canva is great if you are OK with manually manipulating every element of the design, or if you can find a template that has MOST of what you want. Not so good for auto-magically composing a design that matches what you might be envisioning. In Leonardo, I'm not sure that using image guidance is your best route. You might be better off generating the image from a good text prompt using the Phoenix model (especially since you will have text in your image.) Browse the community feed to see if you spot something 'close' to what you're looking for to get some guidance on the settings and prompt details that might achieve that. Another way might be to generate the background, whether that's truly just a background image or generating all the images you want on your cover in leonardo and then take them into Canva and compose them together there and add your text on top within Canva. I'd say you could do some of this in the Leonardo canvas editor, but I don't think it's really beginner-friendly or free-plan friendly. For one-on-one help in Leonardo, I recommend the Leonardo AI Discord. A wonderful mod that goes by Willow B is super helpful and patient. She runs the KZbin channel @WillowsAIAcademy and is one of the kindest people you'll meet.

[@rayfellers]

4 months old and this control panel looks nothing like mine,.

[@excelerator]

Yes, this stuff changes fast! But, if you toggle the Legacy Mode button in the upper right hand corner of the image generation screen, it will look like what showed.

[@kirkhunter88]

Yet ANOTHER unclear tutorial. "Image Guidance" does not exist where you say it does.

[@excelerator]

You're right, and I'm sorry. Leonardo updated the generation screen recently and re-arranged just about everything. But, if you click the toggle on the upper right corner of the generation page to "Legacy" everything will match up to this tutorial. If you stay on the v2 image generation page (the default), image guidance is in the icon that looks like a picture, and it's tucked inside the prompt box. However, there aren't as many options for image guidance in the new layout so you may want to work in the legacy mode. Again, I apologize for the frustration. I will figure out how to note the videos that they're based on what has become the legacy generation interface.

[@kirkhunter88]

@@excelerator Thanks for that clarification. It's one of the reason I don't do online tutorials here... because especially in the AI world, everything changes so quickly that these KZbin tutorials can often be a long trip down a deep rabbit hole.

[@excelerator]

@@kirkhunter88 You're right and, to complicate things, the search results tend to favor older videos that, especially in software and ai, are more likely to be outdated. From a creator standpoint, it was only a few weeks after I published this series that Leonardo revamped the layout of things. For another AI tool I made a video about, the morning I was going to upload the video, the whole platform changed. But, that's not your problem and I'm sorry you're running into these issues, rather than great information.

[@kirkhunter88]

@@excelerator Then I feel you pain!