I mean, i think if you want to perform mass surveilance, you can do it far cheaper and more efficiently via facial recognition, mobile phone surveillance and a variety of different other methods.
If you want reconstruction and training of robotic movement, this is far more appropriate. I believe we're going to see robots being able to "dream" in terms of analysing historical video information on spaces and improving movement and navigation.
So not mass surveilance, but probably there's a future of mass subjugation using robot enforcement.
What is the actual objective of this, is it solving an issue or creating a solution to a problem, that is still to be determined?
It seems like a lot of energy to replicate a lidar mapping system. It's not like you can expect accurate dimensions from this approximate guess work, excluding the expected hallucinations adding to inaccuracy.
Video cameras are much cheaper and easier to use than LIDAR, like anyone can just pull out their phone, take a video and send it to this algorithm to get a reasonable point cloud of the environment. Sure, if you want an exact model of an environment and you have the time and money, LIDAR would give better results, but this is about doing more with less
3D reconstruction of old spaces which no longer exist seems like a clear use case to me. There's loads of old videos of driving down a street in the 80s, or neighborhoods in cities which got replaced.
I can imagine future iterations of this which bring together other stills of the same space at that time to augment the dataset. Then perhaps another pass to fill in gaps with likely missing content based on probability or data from say the same street 10 years later.
It won't be 100% real, but I think it'd be very cool to be able to have a google-street view style experience of areas before google street view existed.
We use drones with RGB cameras for photogrammetry to reconstruct 3D environments with gaussian splatting, which is a manual process and often requires making multiple trips for additional capture to fill in gaps.
Because it's for perceptual use and doesn't require high accuracy, automating with a single-take video would be useful.
One of the key issues of "machine perception" is the inability of machines using standard image sensors to re-create the world accurately.
Lidars are great, and getting smaller, but they still eat a lot of power. (The quest 3 had a lidar on the front[well structured light] and it was mostly not used)
For machines to understand the 3d world, first they need to extract geometry, then isolate those geometries into objects. This method is _a_ way to do that, the first step, extracting 3d points.
The problem with this model is that the points are not actually that well aligned frame to frame. This is why it looks a bit blurry. I assume this is to avoid running out of memory, as you're not quite sure about which points are relevant and need to be kept in memory.
Once you have those points, you need to replace them with simplfied geometry, so that you can workout intersections and junk.
Very interesting paper. I can see street-view using it to perfect the 3D analysing of the photo-video they catch with there google-car. What a wonderfull time we are living in ! Specificaly in the Video to 3D reconstruction.
Every month, a new brick is put in place.Super
> I can see street-view using it to perfect the 3D analysing of the photo-video they catch with there google-car.
Waymo recently announced[1] a World Model that does exactly this: using footage from a single-camera dashcam, it can predict/simulate multiple inputs that would have been sensed by a Waymo vehicle on the same travel path (i.e. multiple camera angles, Lidar cloud, etc). On top of this, the model can be prompted to customize the scenario (adding an elephants or a tornado were the example given)
20 comments
[ 4.3 ms ] story [ 41.5 ms ] threadIf you want reconstruction and training of robotic movement, this is far more appropriate. I believe we're going to see robots being able to "dream" in terms of analysing historical video information on spaces and improving movement and navigation.
So not mass surveilance, but probably there's a future of mass subjugation using robot enforcement.
Relocalisation is the bit thats surveillance-y. But its also crucial for accurate visual only navigation.
> This is a reimplementation of LoGeR; complete code and models will be released upon approval.
I don't understand why it's a reimplementation either?
I would guess it's "research" code anyway so not really usable unless you are an expert.
[1] https://loger-project.github.io/
[2] https://github.com/Junyi42/LoGeR
[3] https://huggingface.co/Junyi42/LoGeR
I can imagine future iterations of this which bring together other stills of the same space at that time to augment the dataset. Then perhaps another pass to fill in gaps with likely missing content based on probability or data from say the same street 10 years later.
It won't be 100% real, but I think it'd be very cool to be able to have a google-street view style experience of areas before google street view existed.
Lidars are great, and getting smaller, but they still eat a lot of power. (The quest 3 had a lidar on the front[well structured light] and it was mostly not used)
For machines to understand the 3d world, first they need to extract geometry, then isolate those geometries into objects. This method is _a_ way to do that, the first step, extracting 3d points.
The problem with this model is that the points are not actually that well aligned frame to frame. This is why it looks a bit blurry. I assume this is to avoid running out of memory, as you're not quite sure about which points are relevant and need to be kept in memory.
Once you have those points, you need to replace them with simplfied geometry, so that you can workout intersections and junk.
[0] https://arstechnica.com/gadgets/2017/09/googles-street-view-...
[1] https://en.wikipedia.org/wiki/Google_Street_View
Waymo recently announced[1] a World Model that does exactly this: using footage from a single-camera dashcam, it can predict/simulate multiple inputs that would have been sensed by a Waymo vehicle on the same travel path (i.e. multiple camera angles, Lidar cloud, etc). On top of this, the model can be prompted to customize the scenario (adding an elephants or a tornado were the example given)
1. https://waymo.com/blog/2026/02/the-waymo-world-model-a-new-f...