- 71
- 241 582
Urban Decoders
United Kingdom
เข้าร่วมเมื่อ 9 เม.ย. 2023
Exploring and Consulting on digital technologies and innovative workflows for Cities and Urbanism. The channel focuses on latest technologies for urban design and architecture with particular focus on digital twins and how they will revolutionize the way we live and design cities. I will also be adding occasional tutorials on related softwares such as the use of Unreal, Cesium, Rhino and Revit for workflows that may benfit those interested in these topics.
Instagram account: urbandecoders
Instagram account: urbandecoders
วีดีโอ
Introduction to Procedural Content Generation (PCG) in Unreal 5
มุมมอง 328หลายเดือนก่อน
Procedural content generation for unreal engine is a major asset introduced from version 5.2 onwards. It allows designers to quickly build complex assets including both landscapes and buildings from basic geometries such as surfaces, splines and meshes which I will be covering over the next few videos. If you are familiar with visual scripting like grasshopper you may find this method very fami...
Best FREE high quality WORLD BUILDING environments in Unreal 5
มุมมอง 1.2K3 หลายเดือนก่อน
World building in unreal engine Allows the creation of vast stunning architectural scenes and environments. However, understanding how these are put together or how they work behind the scenes can be very challenging to say the least. So for this video I will go through 3 high quality scenes you can download in Unreal for free to dive into and break down ideas and assets behind it all. These ex...
AI Sketch to 3D
มุมมอง 4384 หลายเดือนก่อน
Tutorial: th-cam.com/video/NncPqjKOH2w/w-d-xo.html Instagram account: urbandecoders #ai #architecture #generativedesign #parametric #stablediffusion #upscaling #animation #comfui #image #design #generativedesign #painting #sketchtorender #styletransfer #sketchtorender
Sketch with 3D with ComyUI
มุมมอง 1.9K4 หลายเดือนก่อน
The generation of 3D models with AI holds great potential for form finding and designing 3D assets. Particularly the ability being able to quickly sketch an idea and get a 3D representation within an environment can be a very useful workflow especially as this technology becomes more accurate. For this tutorial I will be using TripoSR by stability AI for Fast 3D Object Reconstruction from a Sin...
AI Sketch with Style to Render
มุมมอง 6004 หลายเดือนก่อน
Tutorial: th-cam.com/video/4_37qy9Rfr0/w-d-xo.html Instagram account: urbandecoders #ai #architecture #generativedesign #parametric #stablediffusion #upscaling #animation #comfui #image #design #generativedesign #painting #sketchtorender #styletransfer
Advanced AI Sketch to Image: Style transfer with comfyui and IPadapter
มุมมอง 8K5 หลายเดือนก่อน
The sketch to render has been one of the most popular AI workflows using stable diffusion. The original method mainly relied on control nets however this new workflow I will show you has far more flexibility and control particularly when it comes to styling. This is done through the use of ipadapters which are add-on for using images as prompts. You can use it to copy the style, composition, or...
Super fast Image Generations with SD XL Lightning
มุมมอง 8765 หลายเดือนก่อน
SD XL Lightning is a lightning-fast text-to-image generation model. It can generate high-quality 1024px images in a few steps. It is a model that iterates upon the powerful Stable Diffusion XL (SDXL). while retaining all the standard functionality for image to image, inpainting and upscaling but in a fraction of the time. 00:00 Intro 00:28 Lightning Lora Models 00:58 sdxl lightning demo 01:11 W...
Creative and Sharp Upscaling with AI (latent and ultimate upscaling)
มุมมอง 1.4K5 หลายเดือนก่อน
Upscaling your low resolution images is a workflow you will use quite often in your AI image generation. I will go through one of the most effective workflows that not only sharpens but help generate better and more creative details than the original. You can also easily control how closely to keep to the original content or be more creative in the generation. It is a three part combo that begi...
Morphing Facades
มุมมอง 1.1K6 หลายเดือนก่อน
Facade Optioneering and Morphology with masking and in-painting in comfyUI Tutorial: th-cam.com/video/VtE6ogyIybY/w-d-xo.htmlfeature=shared Instagram account: urbandecoders #ai #architecture #generativedesign #parametric #stablediffusion #animation #comfui #image #design #generativedesign #painting
Best Photorealistic Models with ComfyUI
มุมมอง 3.3K6 หลายเดือนก่อน
I take a Look at two models known for their photo-realism for use with Stable Diffusion XL or sdxl which is great particularly for text to image generation and a huge step up in quality from the standard stable diffusion 1.5 version. The first is RealVisXL V4.0 Juggernaut XL v9. RealVisXL Model: civitai.com/models/139562?modelVersionId=344487 Juggernaut XL Model: civitai.com/models/133005?model...
VR with Google Photorealistic 3D Tiles in Unreal 5
มุมมอง 7976 หลายเดือนก่อน
You can now use google’s photorealitic 3d tiles I combination with VR in unreal engine. I will go through in this video how can combine your own models and use the 3d tiles for context to create immersive VR experiences. You will need to have your unreal engine and PC set up for VR. This includes having your steam VR and occulus apps installed. you can watch my previous video on how to do this....
EASY VR Install for Meta headsets in Unreal Engine 5.3
มุมมอง 5876 หลายเดือนก่อน
The use of XR for architectural visualization and digital twins is becoming increasingly prevalent as the technologies improve and devices get cheaper. In this video I am going to show you how you can set up your meta quest headsets for Unreal engine 5 for you r architectural experiences and apps. I will be doing more videos for VR with architecture and cities so this one is to get the basic se...
Tower Sketch to Render (comfyUI)
มุมมอง 9257 หลายเดือนก่อน
Tower Sketch to Render (comfyUI) Install: th-cam.com/video/nvfmmEUdomc/w-d-xo.htmlfeature=shared Part I: th-cam.com/video/me30yQnh5jU/w-d-xo.htmlfeature=shared Part II: th-cam.com/video/VtE6ogyIybY/w-d-xo.htmlfeature=shared#architecture Part III: th-cam.com/video/oGwTMmWXMjc/w-d-xo.html Instagram account: urbandecoders #ai #architecture #generativedesign #parametric #stablediffus...
Create custom sites with Google Tiles - Quick Streaming (Unreal 5.3)
มุมมอง 2.4K7 หลายเดือนก่อน
In this video I will go through how to clip sites from Google photorealistic tiles that can be streamed into Unreal 5. You will notice that the google tiles are great for aerial views and use in backgrounds, however when you get close up the quality is very low and difficult to insert your own design. This is why the capability to crop and clip out multiple key areas and replace with other tile...
ComfyUI: Intro to Control Net- Sketch to Render (Part III)
มุมมอง 7K7 หลายเดือนก่อน
In this video I will go over how you can develop more advance workflows with comfy UI with control net to allow more structural control based on input images and they are use in popular workflows such as sketch to image AI generation. This follows on from a previous video on the basics of comfy so please take a look at that for more background. Control net is a core extension that I often use a...
Please share your workflow. Thank you!
thanks, what about hatch scale - usually a detail at 1:5 will need a different line density to a section at 1:50 ? is it possible to achieve two different hatch densities on the same wall section in different viewports ?
Hello Brother can you please provide the workflow to download.
Can share link to download workflow.
Please share your workflow. Thank you!
I have inatall contrelnet model but its not showing f16_lineart model not showing
Great video , thanks so much.
What directory do these go into? 2:37
mind sharing the Workflow file?
hi, can Rhino dynamically update a drawing on a layout sheet, as the model is updated, like revit and even sketchup does !
THe ksampler is outputting a black image if I use juggernaut xl and realistic lineart controlnet for sdxl, but it doesn't if I use sd 1.5 models, why?
can I ask how you got rid of the cesium watermark?
i wanna use this to make a large fps map have you tried that yet
Sounds German… recht?
how can i create a structure and use it as reference to the next creation?
add "load picture" node to the workflow. there you can upload the image you created and use it as reference
It doesn’t matter that it doesn’t look good, if you’ve got a procedural city builder this is great as a source of reference.
Is there a way to overlay other data (like traffic for example) on the nearby streets? Do you have a video about that?
yes you can connect to other APIs and read in data such as weather, traffic, IOTs etc. There are some free plugins to do HTTP requests. It is a very interesting workflow, will try to make a simple tutorial for this at some point.
Hi , I face problem with clipseg, the clipseg is installed without errors, but it is not found on the user interface when looking for nodes
is it possible to make the earth flat? and then transfer it to blender? i want to use ivo to create a game
Thank you so so so much looking for this tutorial so long finally got all my answers..... Keep it up really simple easy and quick tutorial not at all boring.........thanks a lot again.....😊😊❤
Hi ! why is this : LogHttp: Warning: Reached threaded request limit (20)
Hi there! How do I get more overlays? I coudn't find any CesiumThreeOverlays in the All folder of the content Drawer.
The current Version of Cesium has deprecated these and they can be found in and COPIED from (not moved) Engine/Plugins/Cesium for Unreal Content/Materials/Deprecated + the Instances folder below this folder. Now to see if it all still works the same way...
HI there! Many thanks for these tutorials. However my CesiumPolygonRasterOverlay is not clipping. Is there anything I am doing wrong? I have repeated the process so many times. Will appreciate a response. Cheers!
Hello there, Never mind! The CesiumPolygonRasterOverlay, should rather be added to Google Photoreaslistic 3D Tileset and not to Cesium3DTileset.
Comfy UI took something simple, and made it complicated. Terrible program
How do you solve the problem of loading Google map tiles in Cesium during loading into the sequencer, as well as during rendering? There's a feature in the Google Photorealistic 3D Tiles section, in the Debug tab: Suspend Update, it stops loading tiles, freezes, but is very limited in terms of map coverage
i have been having issues verifying google console in order to access Google map API. the cards i'm providing are always rejected and that is all i have here in Nigeria. Please help
can i make FBX file with that Polygon plane because i need to use in blender as FBX, help ASAP sir
Hey mate, Brilliant. Can you please make another one which explains clickable events, for instance if one of the pins is clicked that takes us to its own location.
Amaizing!! Thank you soo much!!
can share this workflow, pls
what accent is that?
I can't make the polygon clip, for some reason I can draw the area, I can also make CesiumPolygonRasterOverlay but it doesn't make the cut
in the details panel for the cesiumpolygonrasteroverlay for the Google photorealistic tiles try checking "invert selection" and deselect "exclude selected tiles". Also make sure the polygon is added to the polygons array.
@@Urban_Decoders Thanks it works
Hi, I'm getting the error "Error occurred when executing IPAdapterUnifiedLoader." I've followed all the instructions and read through the comments, but I'm still unable to resolve the issue. I would appreciate your help. Thank you!
The nodes regularly get updated and sometimes causes the nodes to break unfortunately. Could be a few possible fixes. You could try Update All in the manager, and re- download all the IP Adapter files making sure they have correct names and locations. Make sure to have a folder named "ipadapter" inside the "model" folder.
@@Urban_Decoders I have the same issue as everyone else, and I do have an "ipadapter" folder in my models and it's filled with all models I could find, so it has 10 files inside it... But I still get the error: "Error occurred when executing IPAdapterUnifiedLoader: IPAdapter model not found." I don't understand.
@@Urban_Decoders I've finally managed to fix the error. Apparently, I was missing 3 more of the required models in my ipadapter folder, once I've installed them as well, the error went away.
@@itskittyme Can you please send ipadopter model link
@@shamsfx8837 Sorry I only played around with this AI image generator stuff for one day and I have NO idea anymore what any of these comments, even my own comment here, even means >.< so can't help you sorry
Tried it on juggernaut xl lightning but the results were not realistic, sort of like a clay. Can we have a chat for this? Thanks!
Error occurred when executing IPAdapterUnifiedLoader: IPAdapter model not found everything is already updated. I checked the paths. it's nothing ;(
This maybe because you need to download the model but place it not in the Checkpoint folder but in a folder named "ipadapter" If this folder dont exist in your folders, you need to create it.
what Ai?
very good ,thanks
I use realistic vertion V5, I cann't find extendtion DreamArtist?
Thanks for the tutorial. Can I not do this without a Google Earth API subscription?
Yes you can! its much easier now to just do it through a cesium ion account. I made an updated video here using that method: th-cam.com/video/SMxhERRVszU/w-d-xo.html
Theres no point cause it wont look good at street level...and there are obvious disadvantages to using real cities...and to be honest a lot of devs dont want real cities in their fictions...
Depends on what you use Unreal for. For architecture and Digital Twins you want real environments and if you only need aerial views of simplified forms for urban analysis it maybe enough. For more detail closeups, yes you would need aerial drone scans for street details but that is another cost/ service.
@@Urban_Decoders pretty much...good for a flyby in a flight sim....not...much else...don't overlook.that most Devs are telling stories about their own world or universe...and there's other reasons to have a unique city...fight areas...boss areas...etc etc..
Is it suitable for clay render instead of sketch?
yes, same workflow just would recommend using a different control net (maybe depth) for voumes if you don't need details from your input image.
Goddam, I never expected unreal would be this powerful! DEFINITELY WILL USE THIS FOR MY THESIS PRESENTATION IN YEAR TO COME!!
very powerful! You can really tap into Unreal's potential if you start working more wth blueprints/ C++ aswell
thanks Very enlightening
Glad it was helpful!
your sound seems AI... are you real? anyway, nice tutorial!
haha still human, haven't uploaded my conciousness into a machine yet.
Thanks for doing this. Very impressive!
great tutorial! What's the name for the custom_node for Scribble Lines?
You can download the scribble control net from hugging face here and paste into your control net folder: huggingface.co/lllyasviel/ControlNet-v1-1/tree/main
You can also follow the more detailed guide on control nets here that I used: th-cam.com/video/oGwTMmWXMjc/w-d-xo.htmlfeature=shared
Comfi... take a hit of this. Now, make me some stupid house crap in 3D. Oh oh... you inhaled too much. Back of the crackie bits Comf, That's dumb.
Is there any way to export a portion of the city as a mesh and work offline? I'm asking cause I saw many people selling tiles of cities in FBX format on different 3D markets, so I guess it is possible.....but how? Thanks for the videos!
yes you can now download Offline Clips of 3D Tiles with cesiumion. You can read about it here: cesium.com/blog/2024/05/08/cesium-ion-clipping/
Hello there! Great tutorial! I'd like to know if we can bring to UE5 a small area not the entire world? Is that possible?
Yes! you can use cesium cartographic polygons to clip both external and internal areas. I made a tutorial here: th-cam.com/video/SMxhERRVszU/w-d-xo.html
Thx
Thanks for the tutorial, i made same workflow but i got error by executing Ksampler ( Shaps cannot be multiplied) , have you any idea what it could be? thank you
It may be to do that you are mixing SD1.5 models, controlnets, ipa adapters and checkpoints with SDXL ones. Could try double checking that all the models loaded are for the SD 1.5.
@@Urban_Decoders thanks for the hint , i used a wrong Controlnet model. thanks for this tutorial keep it up