Hacker News new | past | comments | ask | show | jobs | submit login
Researchers create open-source platform for Neural Radiance Field development (engineering.berkeley.edu)
80 points by geox on Aug 1, 2023 | hide | past | favorite | 17 comments



I think efforts like this are awesome and I hope to see more in the future, but sometimes I wonder if some for-profit entity will swoop in and yoink all of the hard work these wholesome researchers have put out there for free.

I feel that this happens a lot with well thought out "PhD-level" software. Someone with dollar-sign eyes (who is capable of understanding it) takes the FLOSS and wraps it with a different UI, thereby profiteering off the backs of grad students...


Docs: https://docs.nerf.studio/en/latest/

Src: https://github.com/nerfstudio-project/nerfstudio/

Draft: Neural_Radiance_Fields: https://en.wikipedia.org/wiki/Draft:Neural_Radiance_Fields :

> A Neural Radiance Field (NeRF) is a method based on deep learning for reconstructing a three-dimensional representation of a scene from two-dimensional images. It is a type of generative model that can create 3D scenes from a set of 2D images.[1] The NeRF model can learn the scene geometry, camera poses, and the reflectance properties of objects in a scene, which allows it to render new views of the scene from novel viewpoints.


Nerfstudio is really great and it's nice to see the community consolidate around this wonderful open source project.


The new Photosynth! NeRFs seem mature enough to support meaningful, non-trivial products. I look forward to their development!

https://en.wikipedia.org/wiki/Photosynth


Related:

Localrf – Nerf from casual shaky videos - https://news.ycombinator.com/item?id=36348483 - June 2023 (30 comments)

Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields - https://news.ycombinator.com/item?id=36208942 - June 2023 (47 comments)

HumanRF: High-Fidelity Neural Radiance Fields for Humans in Motion - https://news.ycombinator.com/item?id=35946893 - May 2023 (16 comments)

HDR-NeRF: High Dynamic Range Neural Radiance Fields - https://news.ycombinator.com/item?id=35717106 - April 2023 (36 comments)

HOSNeRF: Dynamic Human-Object-Scene Neural Radiance Fields from a Single Video - https://news.ycombinator.com/item?id=35696328 - April 2023 (1 comment)

Zip-NeRF: Anti-aliased grid-based neural radiance fields [video] - https://news.ycombinator.com/item?id=35569246 - April 2023 (18 comments)

F2-NeRF: Fast Neural Radiance Field Training with Free Camera Trajectories - https://news.ycombinator.com/item?id=35358360 - March 2023 (15 comments)

Nerfstudio: A collaboration friendly studio for NeRFs - https://news.ycombinator.com/item?id=33096449 - Oct 2022 (25 comments, with https://news.ycombinator.com/item?id=33102886 being a particularly good explanation)

MoRF: Morphable radiance fields for multiview neural head modeling - https://news.ycombinator.com/item?id=32237473 - July 2022 (1 comment)

NeRF: An eventual successor for deepfakes? - https://news.ycombinator.com/item?id=31894552 - June 2022 (160 comments)

Block-NeRF: Scalable Large Scene Neural View Synthesis - https://news.ycombinator.com/item?id=30891051 - April 2022 (3 comments)

Nvidia Research Turns 2D Photos into 3D Scenes - https://news.ycombinator.com/item?id=30806925 - March 2022 (228 comments)

Scalable Large Scene Neural View Synthesis - https://news.ycombinator.com/item?id=30299498 - Feb 2022 (12 comments)

DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields - https://news.ycombinator.com/item?id=27658661 - June 2021 (13 comments)

PlenOctrees For Real-time Rendering of Neural Radiance Fields (NeRFs) - https://news.ycombinator.com/item?id=26597419 - March 2021 (31 comments)

Deformable Neural Radiance Fields - https://news.ycombinator.com/item?id=26017783 - Feb 2021 (1 comment)

PixelNeRF Neural Radiance Fields from One or Few Images - https://news.ycombinator.com/item?id=25298426 - Dec 2020 (55 comments)

NeRF in the Wild: reconstructing 3D scenes from internet photography - https://news.ycombinator.com/item?id=24071787 - Aug 2020 (124 comments)

NeRF: Neural Radiance Fields [video] - https://news.ycombinator.com/item?id=22764774 - April 2020 (1 comment)

NeRF: Representing scenes as neural radiance fields for view synthesis - https://news.ycombinator.com/item?id=22637721 - March 2020 (40 comments)


TensoRF – NeRFs from Tensor Decomposition - https://news.ycombinator.com/item?id=30784957 - March 2023 (3 comments)

3D Gaussian Splatting for Real-Time Radiance Field Rendering - https://news.ycombinator.com/item?id=36285374 - June 2023 (2 comments)

This last one is due to be presented at SIGGRAPH 2023, which starts in a few days!


Thank you for the collecting of the linkery, dang. Much appreciated!


> Nerfstudio

I see a trademark cease and desist in your future...


A cease and desist from whom? I tried to search Nerfstudio on https://tmsearch.uspto.gov/bin/gate.exe?f=searchss&state=481... and nothing shows up there.


I'm assuming they mean from the owners of the Nerf brand of foam toys (I think owned by Hasbro nowadays). They could attempt to argue that NerfStudio dilutes the Nerf mark. Not necessarily a frivolous argument and they don't need to show likelihood of confusion in that case.


Trademarks are bound to markets / domains (toys vs computer perception). In my limited understanding, nerf the toys and nerf the computer perception tech, do not serve the same market.


That's not universally true. For most marks, you are right. But for certain particularly "famous" marks, the owner has a potential cause of action for dilution of the owner's mark. Unlike trademark infringement, which requires a showing of likelihood of confusion (which is bolstered by a showing that the two purported mark users serve the same market), trademark dilution does not.


thank you. Makes sense.


It's Nerf or nothin'


I liked the music in the trailer linked there :)


Abstract Fashion Pop by QubeSounds

Cuando Nos Volvamos Encontrar By Corazon Negro


is there any work being done to make nerfs have collision or mesh deformation?




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: