Free AI Art Tool Let's You Build 3D Worlds!
ฝัง
- เผยแพร่เมื่อ 29 พ.ค. 2024
- In this video, I'm exploring several new AI Art tools, including Reimagine from Stability.ai, Face Landmarks, and Nvidia Canvas.
Face Landmarks: huggingface.co/spaces/georgef...
Stable Diffusion Reimagine: clipdrop.co/stable-diffusion-...
Nvidia Canvas: www.nvidia.com/en-us/studio/c...
🛠️ Explore hundreds of AI Tools: futuretools.io/
📰 Weekly Newsletter: www.futuretools.io/newsletter
😊 Discord Community: futuretools.io/discord
🐤 Follow me on Twitter: / mreflow
🐺 My personal blog: mattwolfe.com/
🌯 Buy me a burrito: ko-fi.com/mattwolfe
🍭 My Backgrounds: www.futuretools.io/desktop-ba...
Outro music generated by Mubert mubert.com/render
#AIArt #3dArt #GenerativeArt - วิทยาศาสตร์และเทคโนโลยี
This will be huge for VR eventually. How cool to fast create VR worlds in VR chat !
@Tinselfect can you imagine building worlds from text prompts? Eventually a user interface will be made where you can describe what to put in the world, features, effects, pool tables haha
The user interfaces for AI generation will replace the prompts we use today. We are in the first few iterations of these amazing tools
Your two year old artistic level is pretty impressive, Matt!
This 3D AI tool from NVIDIA is insane. My dream of building myself a 3D game is becoming more reality than ever :)
Matt, your finds and enthusiasm make a great way to start my day. Thanks!
Big fan of your videos. Thank you for posting!
This is amazing, need to experiment with it.
The Nvidia tool is awesome and I have just downloaded it. Thanks for sharing this. Nvidia don't really show up in my social feeds but thats probably not their fault, and its been a busy week of news.
wow theres some great options! I think we have entered an age of digital art that may very well blow entire industry's out of the water! I blv we are heading towards a new frontier where the individual will become as powerful as a small studio.
Another Awesome video packed with great stuff!!!!!!!
Your the hardest TH-camr i view in Ai stuff . thank you 👍
Ahhhh Canvas is a revelation! Thanks dude!!
_You are definitely on top of your game!! You may as well be the AI Tools Journalist, hehe. Cool videos_
Thank you for the video. I have to try more Nvidia tools. I wonder what they will show during this week's conference.
Thanks for another great video. Stable Diffusion Reimagine is insane, I just ran some of my art from PlaygroundAI and WOW! It does have a pro $9 a month option, if you go over the free one.😁 The AI subscription are starting to add up. Thanks for the hidden gem.
Good Videos I subscribed to your channel now. Good work mate
That’s not able to build 3D workds, it’s able to create HDRI images. Big difference, but i appreciate your work a lot!!❤
clickbait
I'm just hoping for a free d-id similar application
You do need an nVidia RTX video card to use their tool, which is unfortunate for those who don't have one (like me).
I wish he mentioned this. Once again, many people are going to be left out in the whole A.I. revolution because of lack of resources. Everything is a subscription if you want to fully use these tools or great hardware to do anything on your PC. I understand why, but it is unfortunate.
Dude I love your videos!
Amazing content thank you!!
Hi thxx for this video is there any tools that allows you to get the same image from different angles?
You can always learn blender3d, that are super cool tutorials out there :))
I've subscribed your channel with this video! Please keep doing these kinds of videos xD!!
Not for long though, since he will be replaced soon by AI thankfully
Could be pretty useful for game jams.
Thanks! This is great!
Thanks for the news👌
Nvidia canvas needs RTX card. I have still gtx 1080ti that have served well. Wonder if this can be done with tablets?
Thank you!
Hey Mat! Any tool with AI tecnology who created a depth pass for any 2d image ( I know stable diffusion has it) but did you know another tool?
Midas
I thought of it, maybe you can create a point cloud and then construct it, like photogrammetry. Lets say you have 4 or more images of the same thing of the same thing but with different views.
I still prefer the way that Bob Ross painted worlds.
Wow. Also I really love all the content you giving us. Amazing times in artificial intelligence.
I am thinking back some 30 years when Silicon Graphics workstations were the bleeding edge in real time rendering.
Anyone remember the ONYX? We had one at the Australian Film and TV school graphics department, it was housed in it's own air conditioned room,
was the size of a SUV 4WD and drew enough power when switched on, that it would dim the lights in the office ha!
Great video !
Can you cover AIs that are available in Poe? It is related to Quora. They have many different AI models available.
Does nvidia canvsa work with GTX cards? Its not in the minimum requirements but wanted to see if anyone has checked.
thank you information your video
Please upload more Matt
I love your enthousiasme. Thanks for showing me all the new stuff that I would me missing otherwise.
daaamn this is nuts!!
Nvidia Canvas Beta isn't available on MAC
Revolutionary❤️
Just an FYI: 360 degree views are not 3D. It's a common misunderstanding. I used to do it a lot too.
Yeah all this stuff needs a long way to go that’s for sure
Nvidia Canvas requires RTX cards?
Nvidia Canvas 🤩 enter Bob Ross impression 🤣
WoW
Nvidia Canvas won't install because it needs an Nvidia RTX GPU which I believe I have but it says it needs to use a WDDM driver and I don't know what that is... so I guess my GPU is too old and if yours is several years old... it probably won't work.... this is really neat software though!
It appears my computer has a GTX and not an RTX which makes a difference...
Hey Matt, make a video on llama/alpaca model. I heard that it is a chatGPT that can be installed locally.
i tought nvdia will just stop with that canvas and there was no new things to add but damn
Got me a little too excited by saying it's 3d matt. Tsk tsk you know what we are really waiting for.
Whait what!? You can draw an picture and turn it to 3d and run with npc or player in it!? Can you interact with it? Or just see it like an background?
you dont how much this helps me, not the expressions but the facial feature alighnment
I crashed chatgpt in a battle of the wits trying to convince it mars evaporated 3 minutes ago and it was outdated from 2021 since this is 2023 now, my source was cartoon network
Is there a mac alternative for nvdia app?
Your sub since you had 1.7k
Wow!
Can you make HDRIs with that?
Canvas looks awesome but sadly no Mac version
Wow
I love to try Canvas but it seems a RTX card it seems. :-(
lol - Stable Diffusion Reimagine is the quintessential opposite of a beautifying Instagram filter.
Nvidia Canvas would be much better if it could output HDR images but it seems like it's standard range for now. Still potentially useful but I imagine most developers will opt to use publicly-available HDR skyboxes from Polyhaven or the asset store.
Should be quite easy to add HDR capabilities and higher resolution. When it can output 8K HDR images, I'll use it for more or less every Unreal Engine (my preferred main tool) project that I make =D
To create your own HDR you could save the image as different times of the day and use each of these in photoshop as the image exposure range and have photoshop create a HDR from them.
You could upscale this using gigapixel AI to get a bit extra from it but there's some AI upscale coming that do a complete re-creation of the image
I give it 3 months before we have text to video tools that produce videos with quality as good as Midjourney 5
blockade labs skybox generator is way ahead on this
Aghhh. Canva panorama requises 6go GPU when I only have 4go... so sad can anybody help me
Lol , light speed innovations
Take picture of a room. Reimagine different furniture. Ai interior design.
Thank you
Saw the video, tried installing Canvas. Did not work. Not really surprised though, it's Nvidia after all.
This is insane! Like imagine you can create your own 3d world then upload it to your VR and can walk around it. Imagine creating a castle you start walking around it
That’s possible in every 3d program... of course it takes time to build 3d models
@@METTI1986LA well yes but I’m saying like put in a prompt and the AI just starts generating it. It would take so much less time
If you do alternative images of yourself The best thing you can actually do is put them through a Toons app And They look much better The AI story I've got my channel called the chocolate wiz that was done like this I used AI art but then I put the result in to a tooning app The chocolate wizard image in the AI story I made it actually me in a top hat but I never actually had a top hat I actually used an app called dawn AI The problem is it doesn't do AI anymore It does preset avatars now which is annoying because they used to allow you to generate what you like but now they seem to be doing this paid avatar thing a bit of a rip off
i gotta say all those ai tools i find kinda boring its like okish sure its the hype right now everyone is talking about as if there was something new to it but anyway what i gotta say im way beyond impressed its like insane at what speed you throw out new videos providing all those hungry for those tools with all the information one person alone propably wouldnt be able to keep track of or even notice .. well you do :) so respect for that level of quality there even tho the topic as such is not for me but i gotta give it to you the quality is the maker :) so i enjoy watching them anyway who knows maybe one day there is a tool in between that actualy turns out to be intresting who knows. so thanks for all the work you put into informing people
this is like the tech in the music video for shelter by porter robinson
Hey Matt, it's awesome to hear that you won't need to share news with us since AI will replace you in probably less than 3 years. Hope you enjoy your time!! :))
When will GPT-4 be available for free to everyone?
I wouldn't count on it being free anytime soon. Subscription is how they make all their money, if it's ever free I would guess it would be after GPT 5 comes out. Which if the past is any indication I would put that release at approximately 2 years from now.
@@KuZiMeiChuan 2025?? thats so long away..
this Nvidia canvas requires Nvidia GPU you forgot to mention that i guess 😢😮
AI will keep me entertained 100%. What it cannot do however, is feed me and pay my bills when there is no job to be paid for.
Like Nvidia, but not MAC users
Darnit. When I saw the title I thought you meant actual 3d objects :( I wouldn't call pano's "3d".
I tried the "Re-imagine" program and it didn't understand the significant features of the original picture.
The generated image wasn't useful at all.
This is a "toy".
Or just an early product lol
@@Letthy_oliverr Yes, I understand this is a "work in progress" so I shouldn't be too harsh. I do graphics work all the time and I can't see how a computer can analyze a picture and, without any guidance at all, recreate a similar, but different, image. ??
I think these AI programs are fascinating but when you get right down to it, they are toys.
I've seen pictures in museums where the artist splattered their brush and that was called "real art".
@@bob456fk6 theyre as useful as you make them. They might seem like toys to you because of what you do in your life but they might save hours or days of work to someone else who does other things...
Theyre not toys, they're tools. If you use them like toys ok but thats your personal use case.
Companies wouldnt invest so much money to create free toys for people lol
@@bob456fk6 gpt 4 is a toy for me, i play wirh it for multiple things, but i saw someone post a tweet where they needed to make some sort of software to control multiple pieces of hardware through a proprietary network interface and senior devs were asking them $5k and two weeks for delivery to make th4 software and the guy and his team made it themselves in 3 hours with gpt 4 doing the programming and it cost them about $1.
All they did was prompt gpt 4 over and over.
Again, theyre tools. How you use them is up to you. But AI is not toys
What I use as a toy just saved tjese people $5k and two weeks of time.
Seems like it's mostly just for casual dabblers that don't have access to local stable diffusion models as it just seems to be img2img with less flexibility to modify the prompt and denoising scale.
why did you say Let's in the title? Are you Felling?
Let's = Contraction of Let Us
Lets = Allow
Felling = misspelling on purpose to draw attention or engagement
:( Canvas needs a RTX card :(
Actually isn't totally free You have so many credits you run out and then you have to open it in another browser to get more
The way this only has 4k views after being up for two hours kills me. How are people not be on this like ants on sugar?!
Stfu and watch
The requirements to use it is a bit much
The stable diffusion reimagine just turned me into a white dude. I guess it wasn't trained on a variety of people
Matt, you're the shit!
Where is the tool to build the 3D worlds??? I saw a program that does 2D images that can be projected on a 3D object to get a panoramic view but that is in no way a '3D world'. Very disappointing.
/ai ai ai
Cool video, could do without the click bait title though.
Nvidia is the goat.
Every single one of these AI tools I have tried thus far give sub par results for the most part. The amount of trial and error that needs to happen to make an amazing piece of art isn't worth the cost, considering the pricing on some of these websites. It could take 50 to 100 attempts to make something decent, which eats up your subscription allotment pretty fast. It also doesn't take long to notice the similarities from one art piece to the next. You can tell it's drawing from the same set of images. Eventually AI will "get there" but it isn't there yet.
[4:27] Happy Hiroshima models
"3D" hehe, more like a 2D mat painting wraped on a 3d Sphere, i was hoping for actual 3d environment generation.
Firsttttt
Mac sucks! Any AI that works pre M1 Or M2? On monterrey?
"""Tool"""… sure
AI art is a quick way to vomit and become psychotic :)
This is nothing. I start out with perfect pieces of art and then AI transforms it into a drawn-by-a-2-yr-old version!
It’s a little misleading to call a 360 panorama a “3D”. It’s a 2D skybox . Nothing 3D about it . Had to sit through 12 minutes of this :/
YOu haven't actually made '3d' spaces before. You use the term 3D way to easily. I don't actually ANY 3D here.
Re-imagine SUCKS! I will never use that again. 5minutes wasted lol. It's hilariously bad. It's like a toddler using photoshop and makes everything look worse. Try it! You'll see. SUCKS
Well the first Stable Diffusion sucked too and some less months later...