News for hackers. Stuff that matters.
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by AndrewKemendo on Monday March 14, 2022 @03:58PM (#30677425)
Honestly this doesn't look any better than what we were doing back in 2016-2017. I'm not sure what's novel here.

This is the only video I could find, but we were doing monocular reconstruction from a limited number of RGB (not depth) images AND doing voxel segmentation on the processing side. https://www.youtube.com/watch?v=nqy44VSWh3g

Even as far back as 2010 people were doing reasonable monocular reconstruction including software like meshroom etc...the whole of TU Munich also under Matthias Niessner has been doing this for a while.

What's novel here?

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by nobbis on Monday March 14, 2022 @06:19PM (#30679196)
Their research doesn't just integrate depth maps into a TSDF - it uses NN's to incorporate surface priors.

I don't recall you having similar real-time meshing functionality in 2016-2017, Andrew. Can you show what you had?

As far as I'm aware, Abound was the first to demo real-time monocular mobile meshing: on Android in early 2017 (e.g. https://www.youtube.com/watch?v=K9CpT-sy7HE), and iOS in early 2018 (e.g. https://twitter.com/nobbis/status/972298968574013440).

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by fxtentacle on Monday March 14, 2022 @06:45PM (#30679504)
This is a paper about a new way of storing/merging 3D data.

The actual 3D reconstruction is so-so, I agree. And they kinda cheat by using ARKit (which uses LIDAR internally) to get good camera poses even if there is little texture.

So the novel part here is that they can immediately merge all the images into a coherent representation of the 3D space, as opposed to first doing bundle adjustment, then doing pairwise depth matching, then doing streak-based depth matching, and then merging the resulting point clouds.

Also, they can use learned 3D shape priors to improve their results. Basically that means "if there is no visible gap, assume the surface is flat". But AFAIK, that's not new.

EDIT: My main criticism of this paper after looking at the source code a bit would be that due to the TSDF, which is like a 3D voxel grid, they need insane amounts of GPU memory, or else the scenes either need to be very small or low resolution. That is most likely also the reason why they reconstruction looks so cartoon-like and is smooth on all corners: They lack the memory to store more high-frequency details.

EDIT2: Mainly, it looks like they managed to reduce the GPU memory consumption of Atlas [1] which is why they can reconstruct larger areas and/or higher resolution. But it's still far less detail than Colmap [2].

[1] https://github.com/magicleap/Atlas

[2] https://colmap.github.io/

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by nobbis on Monday March 14, 2022 @07:12PM (#30679770)
ARKit doesn’t use LiDAR for camera pose tracking.

TSDF memory isn’t an issue since Niessner et al. (2013).

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by onedognight on Tuesday March 15, 2022 @01:05AM (#30681845)
> ARKit doesn’t use LiDAR for camera pose tracking.

It doesn't by default, for power reasons, but it will in a pinch.

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by nobbis on Tuesday March 15, 2022 @08:54AM (#30684619)
Interesting - I’ve never observed it kicking in.

Would love to know in which circumstances it’s used. I assume you work for Apple to know this so understand if you can’t share more.

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by fxtentacle on Tuesday March 15, 2022 @04:28AM (#30682889)
> TSDF memory isn’t an issue since Niessner et al. (2013).

I would strongly disagree. This paper uses TSDF and runs into memory issues. And ATLAS is using TSDF and running into memory issues. So for practical applications, TSDF is still too memory hungry.

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by nobbis on Tuesday March 15, 2022 @08:37AM (#30684504)
Correlation != causation.

Try out our app, Metascan, to see an example of using TSDF with a multi-resolution GPU hashtable that only stores voxel data near surfaces. Or just skim the original voxel hashing paper from 2013 to understand the technique.

Storing voxel data in an array is a lot simpler. So if it’s not the focus of the research, then why would academics engineer something more complex?

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by pj_mukh on Monday March 14, 2022 @04:56PM (#30678185)
Looks like a much better response to white walls/texture less surfaces.
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by dimatura on Monday March 14, 2022 @09:44PM (#30680861)
I haven't read this paper in detail. At a glance, it seems interesting, but not necessarily a huge breakthrough compared to SOTA in monocular 3D reconstruction (which is not a criticism; most papers are incremental and that's fine). That said, what I find with a lot of work on neural techniques applied to 3D reconstruction/VSLAM (a pretty active area in the past few years) is that they're often not significantly more accurate than prior, non-ML techniques, and in the beginnings, often less accurate. On the other hand, they're often more robust. Classical geometry- and optimization- based approaches are fairly brittle, and tend to fail catastrophically or converge to totally implausible solutions in tricky conditions. Learning-based approaches tend to be better at degrading gracefully.
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by jayd16 on Monday March 14, 2022 @09:16PM (#30680694)
Its not a fair fight but your example is much much worse. The mesh reconstruction of the chairs are not suitable for occlusion or physics in your example.
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by closetnerd on Monday March 14, 2022 @04:04PM (#30677495)
Says it’s real-time
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by fxtentacle on Monday March 14, 2022 @06:47PM (#30679529)
I always found ORB-SLAM2 pretty impressive, which can map 3D neighborhoods in realtime while you drive around in a car:

https://www.youtube.com/watch?v=ufvPS5wJAx0

https://www.youtube.com/watch?v=3BrXWH6zRHg

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by nobbis on Monday March 14, 2022 @07:28PM (#30679903)
ORB-SLAM2 is SLAM only (similar to ARKit w/o VIO) so only reconstructs feature points - no surfaces. This research is about surface reconstruction.
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by AndrewKemendo on Monday March 14, 2022 @04:36PM (#30677907)
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by jonas21 on Monday March 14, 2022 @04:38PM (#30677929)
That requires depth input
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by AndrewKemendo on Monday March 14, 2022 @04:54PM (#30678161)
Good point, I don't recall offhand the paper that was the mono-RT one.

At a minimum though 6D.ai and a few others had companies that were selling this as a service at least as far back as 2017.

Re:Real-Time Coherent 3D Reconstruction from Monocular Video by nobbis on Tuesday March 15, 2022 @11:47PM (#30695102)
6D.ai wasn’t released until 2018. You may be thinking of Abound, which you wanted to license in March 2017. I don’t know of anyone else that had real-time meshing in 2017.
Re:Real-Time Coherent 3D Reconstruction from Monocular Video by tintor on Monday March 14, 2022 @04:32PM (#30677868)
Fast enough to be used for mobile robots?