Fantastic! This and the "No CGI is invisible CGI" series made me change my mind on CGI. The issue is CGI has been a crutch for years, and it made audiences think they hate CGI, when in reality *skilled* CGI filling in the world with enough "realism" retained is what we're looking for.

Came over from the Dylan Curious pod. 😊

Wow, fantastic content. Criminally under subscribed.

Incredible Video Ashwin! Thank you - just subscribed!

Gaussian is a distribution function, not an object. Saying "the distribution of gaussians" makes no sense.

What software do they use?

Absolutely love the style of you videos, man. Very clean!!

What is difference between generative design vs topology optimization

I’ve been trying to wrap my brain around wth this is for over a year and now I’m thinking it’s basically physics manipulating ova voxels?

how would you use this in unreal with vr? are there limitations rendering at short distances?

Which Designing software is best for generative designing

This is already nine months old. Do you have any updates?

Awesome thanks for sharing. Living the plugin

But, can it model the real physics, or is it a fancy illusion?

There still hasnt been a very accessible solution for artists who want to make high quality nerfs. Like poly cam is pretty terrible compared to what you see here. I think once there are more easy to use programs for zipnerf and smerf we will see some really amazing stuff

I wonder if you can transform a 3D scene with textures into a gaussian splatting format to make games more optimal without sacrificing too much visual quality

This is awesome. But what if I wanna take the splats to Blender, to insert a 3D statue in the middle of scene? it's possible, or we need some kind of file conversion ?

Would you share the prompt that generated the spinning tesseract, please? :)

Thanks for impressive explanation!!

i tried one of the demo scenes on a google pixel 6 Pro in highend pc quality and got around 3-20 fps in the gardenvase demo and it looked way way better than the smartphone demo quality

the main goal would be having a realtime nerf/smerf/3DGS on a mobile phone to be able to have a realtime photoscamner to get feedback to capture more images on areas that need it that would looks blurry or distorted in the realtime view on a mobile phone.

Fabulous!

that’s music option is amazing, i’m learning drums this will help instead of my having to look to the side at my ipad

samaj nahi i

Great!!! thanks for sharing! Excited to start with this tech!

Amazing video brother!!

Can you make a video explaining how you used runway to reskin your 3D model? Thanks brother!

Does anyone know of a gaussian splat viewer I can put onto a website? So sombody else I can explore my splats?

You literally repeat a good part of your video in your video... 😢

Until you get close to something then it looks like crap

Where does one go to look for talent in this space? Serious question.

i like the part where SMERF is not used anywere and it only exists on github

My God in heaven. 🤯🤯🤯🤯

most VR world limit the number of people that can interact ! if world should be realistic ( not like wii persona ) the power for compute will be to much to be use in mobile context !! Horizon Meta world is still now an awfull empty experience , love IRL ( smellt, touch , ...) !! Google earth VR is really nice ( except resolution ) ... nexstep is hight scan resolution of the world not from satellite or car camera but from human eye level ( house , room, street ...) . in IRL VR or MR is justified for piloting machines that cost a lot in maintenance or fuel, or for avoiding costly travel ( dassault catia 3D), or even avoiding the manufacture of a prototype. thank's to bezier curve research !!

in france TV channel capture in 3D move/body/head and recreate the world after around capture. It like a matrix effect but with camera in the surface of sphere ( 3/4). see in youtube pp garcia "La RÉVOLUTION De La 3D Volumétrique Au JT de 20h ". Technology call 4D view holosys create from french student thesis .It is not like acting in front of a high resolution 180° BIG screen like sony do by compute in real time video on screen refer to the position of the camera and actors !! here we film actors inside a 360 spherical green screen with plenty of cameras !! a kind of VR chat where well design avatar with more expression can be film with many virtual cameras !! after youtube ( video editor) then there ll be place recognition for 3D Designer UV mapping ( Blender photoshop after effect).

this got me super excited but I wish I was smart enough to do all this

Next, AI rgb LED. It can read ur mind and show u what color ur thinking of!

its topology algorithm. stupid youtubers

Great video! I fall more on the side of caution. All these AI integrations seem to offer is 1) having all your sensitive data, conversations, and information even _more_ connected, which means that whenever bad people do figure out how to use it for ill (and they will eventually figure out a way), it will be _really_ bad. And 2) from everything we've heard in the Supreme Court and Congress interrogations of the leaders of Big Tech in the last couple of years, they are _not_ on our side. They're running a business that profits off of using our data to manipulate our thinking, purchasing, and even voting decisions. Why on _Earth_ would I believe them when they say that they won't do the same with this new level of access to our data, all collected together in a nice little basket for them? They've lied before, why wouldn't they do it again? Yes, the tech on its face is _super_ impressive and amazing. But humans are humans, and most humans are selfish, manipulative, and out for _their_ gain, that's it. And I don't hold out the benefit of the doubt for the people running Big Tech, because they're _proven_ that we shouldn't trust them with out data. If Apple moves to implementing AI on all its devices, I'll be moving in the opposite direction, towards a dumber phone. We have no reason to trust them with that much information about us. This all makes me wary, because I _don't_ trust them, and I would honestly rather move back towards living without needing to be constantly connected to a smart phone for everything. It's healthier for us anyway.

Wait a minute, so could this in theory greatly speed up ray tracing ? You render just a few points and then interpolate with gaussian splatting.

Check the pronunciation of "Gaussian" man. You're driving me crazy. 😅

Very comprehensive and insightful.

no creator just editors

Hey! Amazing work! What is the name of the software you're using to relight gs?

I am a VFX artist in Los Angeles who does digital sets, Matte Paintings, and Environments. I am trying to understand Gaussian Splatting from a non white paper stand point. It seems like it takes the capture camera Meta data ...turns it into a point cloud......and when you view it through your camera...it takes the photos or batch of photos that are near by and basically shows you photoprojections from the photo capture? Its basically viewing a flip book of pictures as your moving your camera based on the capture data positions? Forgive my laymen understanding. As a matte painter we do the same thing manually........doing a series of projections based on camera. How far off am I?

lol ... she publicly trashed the company for clout while sitting on the board (here: cset.georgetown.edu/wp-content/uploads/CSET-Decoding-Intentions.pdf). How could a CEO trust and provide information to a board that goes and publicly trashes the company? Ridiculous.

Is this the workings of the Open AI board or intrigues during a later season of Melrose Place?

Didn't knew RC was already doing GS, can you make a video comparing GS results from different softwares?

It seems that the board was courageous in doing the right thing.