Luma AI Dream Machine | DANIEL PIKL

Forget Kling and Sora, there is a new kid on the block Luma Labs AI Dream Machine 🔥🔥🔥
Images: Leonardo AI
music: Deep - Siren’s Cry by Daniel Pikl
#lumaAI #dreammachine #videogen #AI
Hope you guys enjoy this!
👉 If you enjoy this video, please like and share it.
👉 Don't forget to subscribe to this channel for more updates.
👉 Subscribe now: @danielpikl
💢 Stay With Me :
💟 Instagram : / danielpikl
💟 TikTok : / chatbots
💟 Spotify : open.spotify.com/artist/0Y8BS...
💟 X : x.com/danielpikl
💟 Multi web: : linktr.ee/danielpikl

[@marcs7847]

The free plan gives you 30 videos a month... WISH I would have known that before run 18 tests. But now you know!

[@danielpikl]

Yeah, it’s nothing, spent in few minutes 😁

[@birbshid]

It only gives you 10 currently

[@nicholareddy]

After the free runs, how much does it cost to use?

[@danielpikl]

@@nicholareddy 24$/month for 120 videos

[@africannerdmusic]

PER Email address...I got 6 email addresses so I got like 180 a month

[@Siranoxz]

Luma is ready to receive a lot of feedback to help them improve their model.. Which is a lot more welcoming then Sora's closed approach.. In the future we gonna be making high quality movies in 4k/8k/12k/16k quality without having to spend a lot for it lol.. Video games will go the same AI route way by the way.

[@danielpikl]

We’ll see what will happen when Sora came 😁

[@maltheartistme]

the way the girl solves the rubix cube looks like a cartoon

[@danielpikl]

Initial image wasn’t correct but I used it anyways

[@Findin_Dreams]

Impressive! 🔥🎬💯🏆

[@anawakoff]

I used Luma Labs to complete sequences for a fantastic short film I made using various AI tools ( Haiper, Pixverse, gen2 ) and I have to admit that Dream Machine gives promising results.

[@danielpikl]

can you show your video?

[@ralphclark]

Meh. Describing your own artwork as fantastic makes you sound like a self aggrandising jerk

[@Domarius64]

<a href="#" class="seekto" data-time="50">0:50</a> "It's morphin' time!"

[@NightRunner417]

Wow. And here it was just what, a couple of years ago I was laughing myself to tears at an AI Lord Of the Rings video because it was so ridiculously strange and full of errors. This is surely not _perfect,_ but the improvements are coming so quickly now that it's hard to imagine what our world will look like even next year.

[@onlyyoucanstopevil9024]

R.I.P WOKELYWOOD

[@user-gh3di2rc3o]

How will they ever survive this!

[@AndroxineVortex]

​@@user-gh3di2rc3o Scientists HATE him

[@theanonymspysandwich]

These things will go to market with ideologie fully ingrained

[@melon5111]

the ai for sure hates gay people.

[@nihilisticpancake308]

Not really. Now Hollywood can make whatever it wants without artists who can make a fuzz about it. And since they're already a megacorporation. No one can take them down even more since the little guy will be competing with a market oversaturated with AI and viscious dramatubers looking for any upstarting creator to take down over the tiniest mistake and association. So no. WE are the ones screwed. And not to mention culture war BS. Your complaint is Woke stuff so anyone who makes anything slightly progressive will be torn down by right wingers while anyone who isn't full on Woke will get torn down by lefties. So yeah. It might affect Hollywood somewhat. But don't get your hopes too high.

[@xAgentVFX]

<a href="#" class="seekto" data-time="26">0:26</a> is like some X-men movie and shes got superpowers to manipulate reality

[@user-pw4ri3gc8c]

For uncertainty about potential.

[@marcin3136]

Amazing! :)

[@danielpikl]

thanks

[@alonsogarrote8898]

at <a href="#" class="seekto" data-time="92">1:32</a>min you showed a portrait picture, how did you do that? when I uploaded a portrait image it changed it into a square....

[@danielpikl]

Didn’t have problems with image formats for now and I used 16:9, 9:16 and 1:1

[@cfjlkfsjf]

Have you ever wondered what those people inside AI generated pictures or videos are thinking at their surroundings? example people just wondering around a futuristic city ex solar punk picture? I know they aren't really thinking anything, but if it was actually aware of the surroundings and looking on reddit I am like "those guys are so lucky to see that beauty".

[@danielpikl]

Interesting thoughts and one day this could be real 😉

[@user-gh3di2rc3o]

they have rights too. they are my friends.

[@danielpikl]

@@user-gh3di2rc3o 😂👌

[@jichaelmorgan3796]

Cool interesting thoughts

[@marasmusine]

Upload a still frame to Claude 3.5 or GPT4 and ask it what the characters are thinking.

[@voodoochild420ai]

great demo

[@danielpikl]

thank you

[@Guylock]

Awesome trailer. :D

[@danielpikl]

thanks

[@Guylock]

@@danielpikl np.

[@mawot_music]

Where can I find an AI to change things for example in a video that I made before,, remove or add new things??? Thnks

[@danielpikl]

For that you can use Runway runwayml.com/

[@mawot_music]

@@danielpikl thnks

[@v.dickson]

So impressive. I'm on it now on their paid plan and been generating good videos worth the price. Do you know if i can generate a video with a video or just pictures please?

[@danielpikl]

Afaik only image to video and text to video are available right now

[@PSYCHOBEVO]

I've seen people produce interesting results by taking a frame from any video, then generating an AI video with Luma AI by using that single frame as a starting image, then connecting the new AI-generated video seamlessly back into the original video. If you have any video editing software, you can do that pretty easily.

[@danielpikl]

@@PSYCHOBEVO LumaAI added extend feature yesterday so no need to do it manually

[@PSYCHOBEVO]

@@danielpikl Yes, but what I described is how you can generate a video with a video, which is what the original comment was asking about.

[@AI-DreamGameandConcept]

Stably nice for scenery stuff, but the hands seem still the major issue for AI?

[@danielpikl]

Not just hands, there are still lot of “problems” but we’re getting better and better

[@LilPersey]

After this, I kinda wish what we thought was actually what was happening, AI taking over the world 💀 not this.

[@PuchoSprite]

Im literally making videos rugth now lol i use this since sora isn't out yet

[@bcraigcraig4796]

What kind of prompts ,and some tips on Describing a Camera angles ?thanks

[@danielpikl]

I used different combinations for testing purposes, and prompts are connected with images they need to animate, here are some examples: camera slowly zoom out, POW camera zoom out and down, camera pan left, Camera pan up slowly and moving back, Camera rotate and zoom to purple car, Mistical ball moves and rotates in the background, thick smoke around, Camera rotates around colorful car You can try and use "standard" prompts for AI video gen but I can't tell Dream Machine is following them exactly cause I spent free credits in few minutes :D

[@bcraigcraig4796]

@@danielpikl Thanks for sharing great tips and I am also on free plan

[@AI-DreamGameandConcept]

@@danielpikl good, it seems not many tutorials have covered this, you may like to have a detailed video on this :)

[@danielpikl]

@@AI-DreamGameandConcept I’ll keep that in mind 😉

[@choppergirl]

I gave it a photo of me doing some gymnastics on a balance beam, and it pwked out some montrous grrrbage.

[@danielpikl]

@@choppergirl it’s not good and perfect each time

[@choppergirl]

@@danielpiklThere is no intelligence behind it at all, it's just mimicking patterns

[@rogerioaraujo6607]

How do I animate a photo of myself without Luma AI changing the face?

[@danielpikl]

Upload your image in Luma AI Motion and choose smaller Motion Strength so it doesn't deform too much.

[@rogerioaraujo6607]

@@danielpikl This option doesn't appear for me, only "enhance prompt" option. Should I write this in the prompt?

[@danielpikl]

@@rogerioaraujo6607 no you just need to make video with smaller intereference, send me DM if you need more info

[@rogerioaraujo6607]

@@danielpikl Thank you, I’ll send you a DM. Do you prefer email?

[@Comic_Book_Creator]

well this is good, but same time, dont worth to pay .. as looks like lot of need to be done to get it be perfect, not stable , unpredictable

[@FounderGreatest]

What a song??

[@danielpikl]

Song is mine, you can listen here kzbin.info/www/bejne/jYjcnZR6eLeErKMsi=M2S6WmOitbcWvrIK

[@FounderGreatest]

@@danielpikl good song, sir

[@boilinabag]

i notice on luma beta, hands are problem, which will get solved, but i have seen this blur that will be corrected. i've seen football players run thru a opponent like a ghost. but this is coming kids. resolve, googles video, no more need for film sets or post production.

[@djRoyalTee]

@<a href="#" class="seekto" data-time="70">1:10</a> @<a href="#" class="seekto" data-time="85">1:25</a> how is that NOT ™ or ©Infringement? 🤣

[@danielpikl]

What do you mean? All images are mine created with my prompts.

[@djRoyalTee]

@@danielpikl That's not how copyright works... prompts doesn't clear the AI model of copyright. The infringement isn't on YOU, it's on the AI modeler.. the Source of the model is clear where it's from. Again, not on you.

[@TurkThePunkMunkee-kd2zb]

Still pretty much unusable though, right?

[@danielpikl]

If you talking about complete control, then yes.

[@TurkThePunkMunkee-kd2zb]

@@danielpikl I mean the video output, the end result, is still very clearly 'AI generated' as it is still full of glitches and wouldn't pass for real footage.

[@danielpikl]

@@TurkThePunkMunkee-kd2zb in some cases you would be surprised and for certainly you shouldn't know in advance that it was made with AI ;)

[@TurkThePunkMunkee-kd2zb]

@@danielpikl True enough, although, ironically, the really good AI generated video looks like stock footage - the blue parrot for example - I'm sure I'm not alone in sitting and waiting for when my most fantastical and imaginative idea can be turned into equally high quality and realistic looking video images that we can use commercially in our output?

[@danielpikl]

@@TurkThePunkMunkee-kd2zb you're not the only one, we're all waiting patiently for similar services but right now there is none capable to do such things.

[@user-Grandpa]

KZbin kids can now truly generate infinite brainrot lol

[@sonicmax7]

AI junkers can now rip off more people

[@piteshbhanushali1140]

All are text to video ?

[@danielpikl]

No, all are image to video + prompts

[@piteshbhanushali1140]

Prompt please ..

[@danielpikl]

Check my answer above please👆

[@RhapsHayden]

I can animate myself now, yay😊