News for hackers. Stuff that matters.
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by m3kw9 on Wednesday March 29, 2023 @11:40AM (#35358829)
Google streets using this gonna be lit
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by aaroninsf on Wednesday March 29, 2023 @11:51AM (#35358992)
That... is an excellent observation.
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by sorenjan on Wednesday March 29, 2023 @12:20PM (#35359427)
They're already there, although it's probably too computationally costly to do at scale.

https://waymo.com/research/block-nerf/

Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by Lucent on Wednesday March 29, 2023 @01:13PM (#35360180)
How has this not yet been deployed by a startup to unseat three-quarter billion market cap Matterport?
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by fudged71 on Wednesday March 29, 2023 @01:45PM (#35360646)
Luma Labs seems to be doing very well, they just raised some big funding
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by nobbis on Wednesday March 29, 2023 @02:19PM (#35361123)
And Matterport will never see them coming, as their employees are banned from using Luma:

Competitors: No employee, independent contractor, agent, or affiliate of any competing 3-D capture company is permitted to view, access, or use any portion of the Service without express written permission from Luma AI. By viewing, using, or accessing the Service, you represent and warrant that you are not a competitor of Luma AI or any of its affiliates, or acting on behalf of a competitor of Luma AI in using or accessing the Service.

https://lumalabs.ai/legal/tos

(Disclaimer: another banned competitor)

Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by mepian on Wednesday March 29, 2023 @02:39PM (#35361372)
Is this even legal?
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by nobbis on Wednesday March 29, 2023 @02:46PM (#35361474)
IANAL but I assume so. "Employee of a competing company" (whatever that means) isn't a protected class.
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by pierrebai on Wednesday March 29, 2023 @04:10PM (#35362736)
IANAL, but this seems easily circumvented by setting up an independent entity. (It's not an affiliate if it's not owned by the competitor.)
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by fudged71 on Wednesday March 29, 2023 @01:45PM (#35360650)
What a time to be alive!
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by peteforde on Wednesday March 29, 2023 @03:24PM (#35362010)
This seems like it could have exciting implications for VR.

Even after many years, one of my favourite things to do in VR is Google Earth. Yet, as a developer, it's a frustratingly closed platform.

I understand that they don't want me to put steampunk zeppelins into the world, but I so very want that.

Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by azinman2 on Wednesday March 29, 2023 @03:27PM (#35362056)
What vr platform supports it?

I’d love to have some kind of remote robot that live streams to me as it goes around somewhere remote. Yes this has many obvious problems, but to be able to be in Tokyo for a few hours, hear the environment and direct what I want to see would be incredibly exciting to me.

Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by peteforde on Wednesday March 29, 2023 @09:31PM (#35366560)
I primarily work in Unity, but I'm not sure if that answers your question?
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by kleiba on Wednesday March 29, 2023 @03:25PM (#35362041)
I don't understand what exactly I'm looking at, since it's not my area. What is going on here?
Re:F2-NeRF: Fast Neural Radiance Field Training with Free Ca... by riotnrrd on Wednesday March 29, 2023 @05:48PM (#35364236)
Neural radiance fields (NeRFs) is a technique to create models of complex 3d geometry from a set of photo/camera pose pairs. The way it usually works is that the 3d geometry you're interested in (a car, a tree, etc.) exists within a static, bounded space (e.g. a unit cube). This research changes that requirement, allowing you to create larger spaces without the memory and computational drawbacks of "just making the bounded space bigger."