2 ай бұрын
Жыл бұрын
Пікірлер
@christiandebney1989 16 күн бұрын
theres alot of stolen frames from private ryan in there.
@matthallettai 15 күн бұрын
As I mentioned on facebook they're inspired frames. I used img2img to create a few frames because I couldn't get flux to create anything like what I wanted with just text to image. I was more interested in seeing how those images would animate with Runway.
@resetmatrix 21 күн бұрын
thanks, very interesting using ZDepth as as "guide" to new renders
@matthallettai 15 күн бұрын
Glad it was helpful!
@ЕкатеринаИльясова-т3ч 26 күн бұрын
Thank you for the tutorial. Where can I find the model for ControlNet(control_v11fie_sd15_tile)? My dropdown list is empty, and I can't find this model anywhere.
@matthallettai 15 күн бұрын
You can download models required on Hugginface or Civit. Then you have to place them in the appropriate download directory. If you use a Local AI manager like SD Matrix, it has a download menu you can use for automatic placement. Best to start with a startup video for Forge or Automatic1111
@needse1 29 күн бұрын
thanks
@popeopera Ай бұрын
I keep getting "Package Modification Failed" while installing packages....any ideas?? (pip install failed with code2)
@matthallettai 15 күн бұрын
Its a late reply, hopefully you have it figured out. If all your packages get that error then its a problem with Matrix or Windows security perhaps. IF its just one package, that happens during some updates, there's usually a fix posted quickly on Github. Or you have to revert back to a previous version.
@pirlibibi Ай бұрын
Gold tutorial !
@matthallettai 15 күн бұрын
Thank you!
@davida5136 Ай бұрын
DOPE
@davida5136 Ай бұрын
AWESOME
@davida5136 Ай бұрын
SWWEET
@davida5136 Ай бұрын
Awesome!
@davida5136 Ай бұрын
Great!
@helix8847 Ай бұрын
Issue is all I see are the movies that it has taken its inspiration from, That most likely would be because of the limited data.
@matthallettai 15 күн бұрын
I extracted frames from some movies and built a Flux LoRA, the base model of flux has very limited WWII data.
@johnny5132 Ай бұрын
unbelievable!
@chove93 Ай бұрын
thx you
@L30nHbl Ай бұрын
you are genius! thanks alot!
@matthallettai Ай бұрын
Thanks but its Tyson Ibele whos' the genius!
@SouvikKarmakar1 Ай бұрын
Great video , thanks , Wish blender has this addon.
@matthallettai Ай бұрын
There are AI addons for Blender, but I dont know if it has this cool unwrap feature.
@smukkegreen Ай бұрын
Great video. Gotta try this.
@lawebley Ай бұрын
Brilliant stuff!
@RyanDaily Ай бұрын
When installing stability matrix, can you share data with tydiffusion for the sake of drive space?
@matthallettai Ай бұрын
Yes and no. In TyDiffusion you can share a few model directories, but not the main Comfy. The first version of TyDiff had issues with installation and kept having to reinstall so I kept it away from Matrix. TyDiff has very limited model use, so if you have 20GB, I would keep it all separate from now and just use a few checkpoints TyDiff.
@phunkaeg Ай бұрын
Very Awesome Matt!
@pasindumadulupura8462 Ай бұрын
after many vdos Im able to run CN in comfy without errors. Kudos 2u man.
@matthallettai Ай бұрын
Thanks for the comment. Glad I could help.
@3Dsnapper 2 ай бұрын
So I've recently tried shifting from Corona to D5 because of it's AI features. They were awesome but still had some limitations - this is exactly what I was hoping for to try to make up for that. I hope this turns out well 😅Thank you for this tutorial !
@matthallettai Ай бұрын
Thanks for the comment. If your using 3ds max check out TyDiffussion. SD right inside the max window.
@3Dsnapper Ай бұрын
@@matthallettai I've left Max though - Too expensive. Haha. Is there a version / anything similar in Blender ?
@TheMaxvin 2 ай бұрын
Що казати, Fooocus вже має навіть стиль 'Pa Art Ukrainian Folk Art'. І це круто!
@matthallettai 2 ай бұрын
Slava Ukraini!!
@R1PPA-C 2 ай бұрын
Have you worked with the animation side of things yet ? I'm struggling to get the animations to come out like the single images are...the results aren't wildly diffeent but almost like it's using a different model... Also how do you have it setup so that you can see the image as it's generating? mine just goes through the whole process then outputs the final image, I mainly want to see what's happening as the anim is processing as currently I have to wait for the whole sequence to be finalised before I see what the result will look like, thanks :)
@matthallettai 2 ай бұрын
You're always going to have that weird morphing effect with frame by frame SD animation. No matter what tricks you try there no frame is 100% the same as the last. At least currently. I'm sureone out there is working on it. Ai Video you see now is made with video trained models. What we need is a hybrid or controlnet designed for frame by frame denoising img to img. The current tech is animatediff, deform. - see example on this channel. Personally I like SVD but that has little control.
@R1PPA-C 2 ай бұрын
@@matthallettai well the issue I'm having is not the difference in frames but the initial outcome is completely different when doing a single frame with the same settings as when I hit animation. I said not wildly different but sometimes they are... I train a model to be something which I want for each frame but when I go to animate it's like I've used completely different prompts.. I'm lost
@matthallett4126 2 ай бұрын
@R1PPA-C Depending on the complexity of your scene the more interpolation the AI does with what it "sees" the examples you've seen of other animations look smooth because of their simplicity in size and materials. Leaves and grass for example with change dramatically between frames no matter what you do. Small details change so much it's not worth it. Trust me it's not you.
@ChrisCenters 2 ай бұрын
When I first started working in post production, the buzz word of the day was "morph". All my clients would come to me asking for this thing go from this to that, and I would ask, "Okay, how are you seeing the transition from this to that?" They'd all say, "I don't know, maybe it morphs?" not knowing what that means or how it could be done. I wish I had this video to show them back then.
@matthallettai 2 ай бұрын
That's exactly how I describe it to. Morphing. We can't escape it yet.
@omer133 2 ай бұрын
Thank you for the video. What stable diffusion models can you recommend, specifically for interior design and architecture separately?
@matthallettai 2 ай бұрын
Don't bother with any model that claims it's good for it, interiors or architecture. Unless it's a Lora addon to experiment with adding certain looks. My favored checkpoints right now are AlbedoXl 2.1 for exteriors. NightVision. EpicPhotogasm. Real Vision XL some others and spelling is off...I'm away from my PC. Best to download popular XL models that are for photorealism. Portrait examples are OK. And compare them with the XYZ plot script at the bottom of A1111 or Forge. Makes a handy grid for you to compare.
@LudvikKoutnyArt 2 ай бұрын
I believe the technical term for an AI enthusiast is a "proompter" :)
@AB-wf8ek 2 ай бұрын
Not true. Although language is an integral part, with complex node based processes, it's only a fraction of it.
@YansRiegel 2 ай бұрын
Thanks! Great one
@matthallettai 2 ай бұрын
Glad you liked it!
@jhgil2204 2 ай бұрын
I want to know the sequence rendering!
@AB-wf8ek 2 ай бұрын
In order to get animation with temporal consistency, you'll need to use something like ComfyUI, which is a browser based node editor. Just diffusing over individual frames with a plugin like this will look very flickery.
@ramdpshah 2 ай бұрын
Thanks for tutorial🎉🎉
@The-Voxel-design 2 ай бұрын
you are just a genius. Amazing the job you are doing !
@matthallettai 2 ай бұрын
Ah thanks man! Thats so kind of you.
@ivanibanez1273 2 ай бұрын
Finally!!
@matthallettai 2 ай бұрын
I hope you found it useful.
@shinonyx 2 ай бұрын
Interesting, i just started using SD for this purpose, i didn't know there was a 1 click method.
@matthallettai 2 ай бұрын
I should edit this video to say the Adetailer has these features now, you don't have to use the same one I'm using. I hope you get it working!
@mosambielal6700 2 ай бұрын
Can you please guide about how to do this exact same thing for video?
@matthallettai 2 ай бұрын
Doesn't work for video, each frame ends up being slightly different and you get massive flickering. I use Waifu2x for video by Aaron Feng.
@Objektiv_J 3 ай бұрын
Awesome! Thanks for the video, Matt.
@Man-Tee 3 ай бұрын
Hi, sir. thank for your tutorior, that's so great and help me a lots, but could you help me a bit for this, my "upcaler" is not same like yours, and i tried to dowload these file to upscale (ultrasharp) but i don't know where place it. could you help me this? thanks advaced!
@matthallettai 2 ай бұрын
A little late for a reply. Most of the upscale models you download go into models/ESRGAN
@franciscoklein2249 3 ай бұрын
thanks matt for the tutorial in spanish
@matthallettai 3 ай бұрын
You're welcome, mate. You're the first one to comment on that option. Thank you.
@3DArchiRender 3 ай бұрын
Great tutorial! Can you do this with comfyui as well, please.
@matthallettai 2 ай бұрын
I don't use comfy very often. I have so many nodes installed that its bloated and so confusing for a causal user such as myself. I only use it when something new comes out that I can test and compare it to Forge. But I'll put it in the queue!
@johnny5132 3 ай бұрын
fantastic thanks Matt
@thanhvule4065 4 ай бұрын
thank you so much Sir!!!!
@matthallettai 3 ай бұрын
You're welcome!
@svsguru2000 4 ай бұрын
is it possible to feed it an empty room and have it fill it with furniture?
@matthallettai 3 ай бұрын
Totally. Follow the same steps without any furniture, but raise the "Starting Control Step" to 0.1 and the "End Step" to 0.8. You may have to adjust those, they're just off the top of my head. You'll need to describe the room as a positive prompt very clearly. "Photo of a modern interior, with a sofa in the centre of a large room with a rug, and a window in the background" something like that.
@marcoyaca 4 ай бұрын
genio! gracias por tan buen video
@Beauty.and.FashionPhotographer 4 ай бұрын
how would you go about adding skinpores in a face that has merely plastic face skin , so no noise. in an image from midjourney that is great besides that skin in the face which is fake only because of the that non existent skin. Hair in this photo , its 100% real photo, but skin pores on face is like a comic animee illustration. i have been pondering and testing for weeks now, and none of the people here on youtube, where it worked for them in their own videos, then after it actually did not work at all, when i was doing this. its a niche thing. but you are doing the same only with architecture and products. would the process be similar to add new healthy real skin pores in a face around 2000 x2000 pixels , where there is no face skin foundation at all, so no noise in that face skin, no pre-existing bad skin on that face neither ?
@matthallettai 3 ай бұрын
I would check out my upscaling and enhancement video. It talks about adding detail, and it works well at that resolution. I would also look for a skin LoRA model on CivitAI. You add that to your LoRA directory, and add it in the prompts. There will be instructions how to use it on the Civitai page.
@erikwest7299 5 ай бұрын
sub'd... looks like you almost tripled your subs since you made this video ✌
@matthallettai 5 ай бұрын
Thanks Erik, ya its a slow climb, especially when I don't have any free time to make videos!
@L30nHbl 5 ай бұрын
thank you! great stuff!
@matthallettai 5 ай бұрын
You're welcome , thank for the comment.
@joshmyers1413 5 ай бұрын
Can the output images be used for commercial use?
@matthallettai 5 ай бұрын
Yes. there's no copy right laws for AI art as far as I know.
@ignaciocasotto5315 5 ай бұрын
Hi!! Amazing tutorial and results. I also work on Real State Visualization at "Almost Real Viz" study. I have a RTX 3070, 128gb ram and 5950x, with a same size image that you, and same settings/parameters, on my computer it take 24hs to finish the process.... Do you know what can I doing wrong or if take too long just because my computer?
@ignaciocasotto5315 5 ай бұрын
I had something wrong on SD configuration, but still working to slow, take 2 hours for me...
@matthallettai 5 ай бұрын
Its not your computer. All the processing is done on the GPU. Go back to basics and check all your settings, things that can make it slow are sampling steps and resolution. Try using "NEAREST" upscaler for testing, since that is the fastest. Run all of this on a SSD if you can... Thats all I can think of for now.
@EvlixManiac 6 ай бұрын
you can make discounts tho, cuz man i live in a 3rd world country and your tutorials are amazing
@matthallettai 5 ай бұрын
I've done that before, email me.
@danielmorgado1613 6 ай бұрын
Hey, Matt. I hope youre doing well! Thanks for sharing this, your time invested in this tutorials is very valuable. Ill definately buy some of your courses. Everything is very well explained and its clear that youve invested a lot of time testing and understading all this new tec. Thanks, again!!
@matthallettai 5 ай бұрын
Wow, thanks Daniel! Thats so kind and thoughtful, someone might think you're a bot I paid to write reviews ;)
@danielmorgado1613 5 ай бұрын
@@matthallettai Im human! I swear! haha
@mixocg 6 ай бұрын
hey Matt, thanks for the video! Which GPU are you using in this tutorial? i have 1080 8GB and it's so slow, planning for upgrade in the near future.
@matthallettai 5 ай бұрын
I'm using a 4090. I use it mostly for GPU rendering. 1080 is pretty old now. I've tested it on my old 2080ti and its pretty good for SD 1.5. I'm sure you could find those used for cheap. Look for a GPU with as many CUDA cores as you can afford and at least 12GB of VRAM You can also try Runpod or other online GPU rental services. Try sticking to 512 or 768 resolutions when upscaling and samples at 20.