Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

VR Is Not Dead

yadin.com

Post reply on HN
Print view View original

61–68 of 68 posts

Re: VR Is Not Dead

#61

Post by jerkstate »

echelon wrote: post #7
jerkstate wrote: post #5

With AI, VR is even more promising. I have been working on a Gaussian splat renderer for the Quest 3, and by having Claude and ChatGPT read state-of-the-art papers, I have been able to build a training and rendering pipeline that is getting >50 fps for large indoor scenes on the Quest 3. I started with an (AI-driven) port of a desktop renderer, which got less than 1 fps, but I've integrated both training and renderin…

What's your take on WorldLabs and Apple's splat models? Are there other open source alternatives? How would editing work? Do you think these will win over video world models like Genie? Have you played with DiamondWM and other open source video world models?

My understanding is that those models create gaussian splats from a text prompt, kinda like a 3d version of nano banana. I'm not doing that (yet), what I'm doing is creating splats from a set of photos - aka "splat training" and then rendering the splat as a static (working on dynamism) on the Quest headset. This is pretty well-worn territory with a lot of good implementations, but I have my own implementation of a trainer in C++/CUDA (originally based in SpeedySplat, which was written in Python, but now completely rewritten and not much of SpeedySplat is left) and renderer in C++/OpenXR for the Quest (originally based on a LLM-made port of 3DGS.cpp to OpenXR, but 100% rewritten now), and I can easily integrate techniques from research.

Re: VR Is Not Dead

#62

Post by jerkstate »

guyomes wrote: post #27
jerkstate wrote: post #5

With AI, VR is even more promising. I have been working on a Gaussian splat renderer for the Quest 3, and by having Claude and ChatGPT read state-of-the-art papers, I have been able to build a training and rendering pipeline that is getting >50 fps for large indoor scenes on the Quest 3. I started with an (AI-driven) port of a desktop renderer, which got less than 1 fps, but I've integrated both training and renderin…

You might be interested in a new experimental 3D scene learning and rendering approach called Radiant foam [1], which is supposed to be better suited for GPUs that don't have hardware ray tracing acceleration. [1]: https://radfoam.github.io/

Cool! I'll definitely check it out. The great thing about LLMs is I can probably have a trainer and renderer using this technology up and running for my platform in a day or two, OR I can just pick and choose parts that would work well for my implementation and merge them in.

Re: VR Is Not Dead

#63

Post by socalgal2 »

canadaduane wrote: post #28
socalgal2 wrote: post #22

Earlier quoted context omitted.

I feel that's like saying "I can get just as fully immersed in a book so who needs movies?" They're different experiences. I don't need Tetris or PacMan in VR. Conversely, Half-Life 1/2 etc are not remotely intense as Half-Life Alyx. In the first 2 you're watching a movie. In the later you're in the world of Half-Life

The frequency of choosing to go out to the movies is also about how often I think "I wish I could do this in VR". Examples: - Before going on a trip, pre-visiting the destination in Google Earth with VR is very spatially informative & makes directional intuition memorable upon arrival at the real world destination. - Virtual role-play with environmental cues that cause make-believe to be ever more real. But most peop…

people watch more movies (video) than read books. TikTok, youtube, netflix, TV

Re: VR Is Not Dead

#64

Post by socalgal2 »

dminik wrote: post #29
socalgal2 wrote: post #22

Earlier quoted context omitted.

I feel that's like saying "I can get just as fully immersed in a book so who needs movies?" They're different experiences. I don't need Tetris or PacMan in VR. Conversely, Half-Life 1/2 etc are not remotely intense as Half-Life Alyx. In the first 2 you're watching a movie. In the later you're in the world of Half-Life

The giant advantage regular games have is that I've yet to smash my hand into a wall playing them. I think that the relatively low living space area for most of the world is a huge strain on VR adoption.

I think that will change when VR is like Striking Vipers.

Re: VR Is Not Dead

#65

Post by Ancapistani »

sunaurus wrote: post #37
manbash wrote: post #23

Earlier quoted context omitted.

As someone who has both interest in VR and racing sims (and other sims), and tried VR and loved it, I am genuinely NOT INTERESTED in owning a headset for the obvious health reasons that come with using one. There is no way this ever can be close to safe for your health than, say, not using it.

I'm really curious what you consider to be the obvious health reasons - it's far from obvious for me.

Me, too.

When I was playing standalone VR (Quest 2/3), I was pretty much always sore from moving around while playing. I moved to PCVR a few years ago, and I still move significantly more just from twisting around to look behind me (combat flight sims).

I can’t see any way that it’s not at least better than a monitor.

Re: VR Is Not Dead

#66

Post by jerkstate »

jerkstate wrote: post #60
paldepind2 wrote: post #35

Earlier quoted context omitted.

Sorry if this is a basic question, but what's you workflow for feeding the papers into the LLM and getting the implementation done? The coding agents that I've used are not able to read PDFs, so I've been wondering how to do it.

this is actually a great question - I just extract the text with PyPDF, but did a brief search on the functionality I'd like to have (convert math equations to LaTeX, extract images, reformat in markdown, extract data from charts) and it looks like there are a couple of promising Python libs like Docling and Marker.. I should really improve this part of my workflow.

after looking into it for a little while, Docling and Marker work pretty well but are very slow. I haven't found anything else that extracts math suitably. It takes 10+ minutes per pdf, so I'm going to run it on a batch of these papers overnight and create my own little gaussian splatting RAG database. It's really too bad PDF is so terrible.

Re: VR Is Not Dead

#67

Post by j45 »

krapp wrote: post #52
j45 wrote: post #39

VR isn't dead. It's not alive either. It's just not it's time yet for mass adoption, however it might be. VR/AR continues to have an increasing footprint of usage and adoption as the technologies evolve. This has been the same cycle for over 20 years. Once we get past 4K displays to 8K at a reasonable cost level, it again will introduce a new audience to VR/AR, and a larger and larger existing audience will sit here…

I don't think VR will ever reach mass adoption, I don't think there's any reason for it to. It isn't sufficiently superior to screens and keyboard/mouse or phones to warrant it. I liken the hype around VR to the hype around LLMs - it's good at some things, but not everything.

I don't know that it will either, I'm equally not sure if it's relevant that it must reach mass adoption to be relevant.

Re: VR Is Not Dead

#68

Post by slau »

krapp wrote: post #47
slau wrote: post #30

The biggest impediment for VR is the fundamentally asocial nature of it. If there are fewer headsets in a room than there are people, it’s going to be awkward for at least one person. Trying to help someone debug something in their headset without me being able to see what they see is a problem (granted, this could be solved by software). Having to share headsets sucks. You have to faff with head straps, adjust IPD,…

Are video game consoles "fundamentally asocial" because there are likely fewer controllers than people in a household? Are computers, because they only have one mouse and keyboard? The existence of VR chat suggests it handles social gaming just fine. I can think of a lot of impediments to VR (the weight of the headset and vertigo being the biggest) but needing everyone in a room to share a single headset at the same…

Apologies if this wasn’t clear, I thought it was obvious: you have something sitting on your face, isolating you physically, visually, physically, and emotionally.

When I’m playing on my couch with my wife, when something happens on screen we still look at each other and laugh—regardless of whether it’s a single player game or not. There’s eye contact.

If I’m engrossed in a game of RL in my office, I can still look down at my dog when she comes and boops me. There’s eye contact.

Virtual reality, for all its qualities and ability to let you be digitally present with people online or also in VR, is physically isolating users from the people who are physically nearby.

Post reply on HN