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

Atlas: End-to-End 3D Scene Reconstruction from Posed Images

github.com

Post reply on HN
Print view View original

1–10 of 26 posts

Re: Atlas: End-to-End 3D Scene Reconstruction from Posed Images

#2

Post by kanobo »

On a tangential thought, it's interesting to me that a company (magicleap) that has raised several billion dollars generates so little value compared to other companies its size that this is the most notable output from them in a year and I thought it was a phd project until I looked at the project owner. Anyways, it's a very interesting project and thanks for sharing.

Re: Atlas: End-to-End 3D Scene Reconstruction from Posed Images

#3

Post by cs702 »

From the folks at Magic Leap. It looks remarkably good to me.

The video at https://www.youtube.com/watch?v=9NOPcOGV6nU&feature=youtu.be is worth watching, especially the parts showing how the model gradually constructs and improves a labeled 3D mesh of a live room as it is fed more visual data by walking around the room.

--

On a related note, Magic Leap has been trying to find a buyer for the business for several months now:

https://www.roadtovr.com/report-magic-leap-buyer-sale/

https://www.bloomberg.com/news/articles/2020-03-11/augmented...

Re: Atlas: End-to-End 3D Scene Reconstruction from Posed Images

#4

Post by rayuela »

kanobo wrote: post #2

On a tangential thought, it's interesting to me that a company (magicleap) that has raised several billion dollars generates so little value compared to other companies its size that this is the most notable output from them in a year and I thought it was a phd project until I looked at the project owner. Anyways, it's a very interesting project and thanks for sharing.

Yeah I have to agree. If this were a PhD thesis it would certainly deserve some praise, but given that this is the most exciting thing to come out of magic leap in years just barely puts them on par with SOTA...well I would be pretty pissed if I was an investor in them.

Re: Atlas: End-to-End 3D Scene Reconstruction from Posed Images

#5

Post by toomuchtodo »

For anyone with domain knowledge, how applicable is Google's NeRF work here in comparison? Is there any overlap?

https://nerf-w.github.io/

https://news.ycombinator.com/item?id=24071787

EDIT: @bitl: Tremendous, thanks for the reply. Would be amazing to be able to build these scenes just by walking around scanning a room with your mobile phone while it records video for processing the frames into scenes (especially considering mobile platforms with a depth sensor for enrichment of the collected data).

Re: Atlas: End-to-End 3D Scene Reconstruction from Posed Images

#6

Post by bitL »

toomuchtodo wrote: post #5

For anyone with domain knowledge, how applicable is Google's NeRF work here in comparison? Is there any overlap? https://nerf-w.github.io/ https://news.ycombinator.com/item?id=24071787 EDIT: @bitl: Tremendous, thanks for the reply. Would be amazing to be able to build these scenes just by walking around scanning a room with your mobile phone while it records video for processing the frames into scenes (especially con…

NeRF has a potential to make all those classical methods obsolete, though it requires many input images and I am not sure how it handles rolling shutter and other distortions.

Re: Atlas: End-to-End 3D Scene Reconstruction from Posed Images

#7

Post by caiobegotti »

cs702 wrote: post #3

From the folks at Magic Leap. It looks remarkably good to me. The video at https://www.youtube.com/watch?v=9NOPcOGV6nU&feature=youtu.be is worth watching, especially the parts showing how the model gradually constructs and improves a labeled 3D mesh of a live room as it is fed more visual data by walking around the room. -- On a related note, Magic Leap has been trying to find a buyer for the business for several mon…

I have no experience in this field at all and they note on the video that the sequence shown was not realtime but I wonder how far we're from having something like this running in realtime or how "realtime" it could be given fancy hardware to be used in the wild?

Re: Atlas: End-to-End 3D Scene Reconstruction from Posed Images

#9

Post by nestorD »

toomuchtodo wrote: post #5

For anyone with domain knowledge, how applicable is Google's NeRF work here in comparison? Is there any overlap? https://nerf-w.github.io/ https://news.ycombinator.com/item?id=24071787 EDIT: @bitl: Tremendous, thanks for the reply. Would be amazing to be able to build these scenes just by walking around scanning a room with your mobile phone while it records video for processing the frames into scenes (especially con…

By default NeRF does not produce a mesh (but one could use marching cubes as does Atlas) and it requires training a neural network for each scene whereas Atlas (as far as I understand it) uses pretrained network to process new scenes.

NeRF would probably produce a much better final result but the Atlas approach (no need to train something from scratch) is the only one that can hope to be run in real time which is vital for some application.

Post reply on HN