What I love about gaussian splats is the way they degrade - instead of a hard cutoff or LoD changing spheres into cubes etc., they get increasingly "dreamy" - the basic idea is still there, just less detailed.
If you navigate closer to the trees, things around you become blurry - as if the very fabric of reality unraveled.
danybittel 5 minutes ago [-]
I like that they are sort of between a photo and a 3D model. Nothing quite like it.
scrumper 58 minutes ago [-]
I don't know anything about them but it's a cool effect. At least on this strawberry, you're not zooming in but rather traveling closer. I don't see the increasing (made up) detail you'd expect from a zoom, we sort of pop through the skin into an invented interior.
chimpanzee2 51 minutes ago [-]
Just wow!
As I scrolled through the website, I was even more impressed with this one though!
Can you elaborate why you chose slang-splat over let's say Lichtfeld studio?
How does it compare to the other splat-training tools?
danybittel 1 minutes ago [-]
I usually use PostShot.. but the quality was not very good. I want to try LichtFeld but my Graphics Card has too little memory.. so I reached out on twitter, some people ran tests and Mykhailo got some better quality out of it so I took his training. You can d/l the COLMAP dataset for free and try yourself.
(I'm ... still not sure what I'm looking at on TFA, and whether or not my browser configuration fails to fully present the site as intended.... OK, if all you're seeing is a blurred image of a strawberry, yes, you'll want to enable a bunch of JS resources. I'm using uMatrix, several hosts must be enabled.)
Lots of translucent blobs composited to produce the appearance of a strawberry.
There is no mesh or model. The visual surface of the strawberry could be made up of blobs spaced far apart physically and not where the surface appears to be.
This is why they are called radiance fields, they model the light not the geometry.
Practically the blobs positions/rotations can be constrained to better physically match the geometry of a strawberry.
KeplerBoy 1 hours ago [-]
I'm not sure i agree. The blobs are exactly where the surface appear to be because they are constrained by multiple viewing angles.
Otherwise the splat would fall apart as soon as the viewing angle is changed slightly (Which it absolutely does in many examples on supersplat, you cannot really create an out of distribution view with 3GS, it's not magic)
jaccola 24 minutes ago [-]
Yes, my statement was loose. The blob doesn’t really have a position since it is theoretically an infinite distribution in 3 space.
It has a mean, and that mean doesn’t have to lie on the surface, consider the case where the mean is deep inside the strawberry but its spike contributes to the surface appearance (e.g a seed could be represented this way, or it could be represented by a small well-oriented blob on the surface, the optimiser doesn’t care)
Looking at all the outdoor scenes that you can walk around, I wonder how long until we start seeing this in places like Google Maps/Earth, as a replacement for the low-res 3D renderings we have now.
I guess the number of samples required to generate a GS is the constraint now, but maybe that will get solved.
rendering them is also non trivial from what i understand, since the millions of indivudal point positions need to be recalculated all the time while moving
evrimoztamur 40 minutes ago [-]
There is a faint sensation of translucency, I wonder if that's an artefact of the process, or if it's the actual optics of the surface layer if the strawberry...
danybittel 22 minutes ago [-]
It's an artefact unfortunately. Gaussian splats have no concept of refraction and have a hard time dealing with reflection. Highlights, usually so so. Something I always wrangle.
vanderZwan 28 minutes ago [-]
Well, at certain angles you actually get very noticeable gaps in the strawberry that are visible when rotating it, but almost invisible when static. I think it's mostly due to that.
ovenchips 1 hours ago [-]
I built PlayCanvas in 2011 to power video games. Here we are in 2026 and it's powering strawberries.
The filesize of a 3d animated splat is seemingly very small, and the method enables ~arbitrary FPS. But it seems the setup required to record it is still huge and expensive, which limits its usefulness.
Even with that there are some interesting use cases, eg. I'd love to be able to watch concerts this way, and freely move around the stage and crowd from any angle.
idoco 36 minutes ago [-]
Great video. I was about to share it here myself.
josh-wrale 2 hours ago [-]
Someone: Please combine microscopy with gaussian splatting.
jcattle 2 hours ago [-]
Don't know if this would be in your wheelhouse, but for very nice macro splats, check out the work by Dany Bittel: https://danybittel.ch/macro.html
Edit: I completely missed that this was posted by him (:
Kalendermann 2 hours ago [-]
This was also my first thought when I zoomed in into the strawberry. I wonder if you can achieve a microscope like effect with a more suitable setup. E.g. better lighting, zoom, lens, etc.
danybittel 2 hours ago [-]
I have done 2x macro (an ant).. and want to try 5x.. but as you get closer, the depth of field becomes really shallow. You can do focus stacking but you risk that the individual areas in focus are less ideal aligned and the tracking can't make any sense out of the geometry anymore.
My intuition is that in theory focus stacking should not be necessary as preprocessing step for 3dgs (or photogrammetry). Does anyone know if there is any recent developments in this regard?
Focus stacking generally is not perfect process and can lead to artifacts/errors and I'd imagine those can then compound when stacked images are used for 3dgs. Also the image focus actually provides some depth data in itself that could be useful?
My take.. at a macro scale, the dof is usually so small, that it's hard to get a reliably track. So you'd need some sort of way to tell that these stacked photos belong into a series, and then you sort of are doing focus stacking :-)
I do think the alignment algorithm could be improved. Maybe the approaches I linked could be used to make a much more robust focus stacking algorithm, that also corrects for 3D geometry. That would be really cool!
KeplerBoy 1 hours ago [-]
If you don't focus stack and try to train on partially unfocused images, the optimizer will try to match the rendered view to be also partially unfocused.
You would have to mask out the blurry areas for each image. I guess one could just implement a feature where the optimizer only optimizes gaussians within the sharp distances relative to the camera.
zokier 27 minutes ago [-]
Other way of looking at the question is if you could make focus stacking better by using the full multi-view dataset? Afaik focus stacking essentially does depth estimation so it seems like multiple views would help with that.
Another way would be some kind of 4d GS where one dimension is the focus distance. But I'd guess the renders would inherently have shallow depth then, which is less useful usually.
arnabdey0503 19 minutes ago [-]
Very nice. did you segment the images before?
vessenes 2 hours ago [-]
Dany, this is so cool.
I'm wondering if the splat community has decided this paper is valuable -- https://github.com/fraunhoferhhi/Self-Organizing-Gaussians -- looking at all the detail in the strawberry splat made me wonder how small one can get the download, and what the current state of the art is for compression.
danybittel 2 hours ago [-]
Thanks! We have two compressed formats, the sog by PlayCanvas and spz with sparkjs. Both now support LODs and compress really well.
bestouff 51 minutes ago [-]
This doesn't work at all for me (Linux desktop, tried with Firefox and Chrome). I only see "fullscreen-extended blurry thumbnails" of the splats.
probably_wrong 39 minutes ago [-]
Works for me with Firefox 140.10.2esr on Devuan 5.
slimbuck 48 minutes ago [-]
If you can, please check any output/errors on the developer console?
carlos-menezes 2 hours ago [-]
You might want to throw that one away :)
danybittel 2 hours ago [-]
It is in fact still on it's mount and slowly rotting / molding.. for a second capture :-)
p0w3n3d 2 hours ago [-]
Yup. It's rotten on the other side. Or maybe lead poisoned
gobdovan 2 hours ago [-]
Sorry if I fell for Poe's law, but just for clarity, the strawberry's rotten underside is most likely missing splats in the rendering.
p0w3n3d 1 hours ago [-]
Of course it is. That's where the joke comes from...
Edit. TIL Poe's law
dudefeliciano 2 hours ago [-]
not sure if that's a joke but i think that's just the effect of the strawberry being placed on a glass/plastic surface to be filmed from underneath
mgaunard 2 hours ago [-]
What happened to the bottom of that poor strawberry?
danybittel 2 hours ago [-]
It was mounted at the bottom.. and I can't quite reach it with my camera. Might have to try some two pass way.
CatMustard 2 hours ago [-]
I wonder would a good, sharp needle and thread make for good mounting for a soft object like this? Thread the needle, pass it right through the strawberry and the secure the thread on something above and below. As long as the strawberry doesn't slide down the thread (hopefully a strawberry is light enough friction would hold it in place!)
Anyway, very cool splat, fair play
gobdovan 2 hours ago [-]
Mount it on a needle/skewer, it should let you capture it in one pass.
4gotunameagain 2 hours ago [-]
Assuming that the person that did this has not tried that. If you look at the setup photos, the grape is resting on a couple of nails. This suggests that many different things have been tried.
1 hours ago [-]
josh-wrale 1 hours ago [-]
How about green screen + rotate the strawberry on a skewer
gobdovan 2 hours ago [-]
Gaussian splat casualty. The bottom looks like partially missing from the reconstruction.
timonoko 2 hours ago [-]
When you cut the splat in half, result is either fuzzy fog or sort-of fibrous crystals. As depicted here.
a1o 2 hours ago [-]
Can you show the setup?
(Can we do a Gaussian Splat of the setup of the photograph for the Gaussian Splat of the Strawberry?)
THAT IS AMAZING!! Seriously, impressive work! Thanks for the link.
galsapir 2 hours ago [-]
From the link: "Shot from 90 perspectives, 88 focus stacked images each. Nikon Z8, full frame, f/7.1, exposure 1/160, ISO 100, Laowa 180mm macro lens, with LED light and bluescreen."
Insane!
danybittel 2 hours ago [-]
And it only takes 20 Minutes to shoot all 7920 photos, the Z8 is crazy fast.
dudefeliciano 10 minutes ago [-]
i made some decent splats (admittedly much lower quality than this) by taking video with my iphone13 mini and then chopping it up into individual frames via ffmpeg
voidUpdate 2 hours ago [-]
Gaussian splats look really good from a distance, but as soon as you zoom in, they really fall off a cliff :/
classified 20 minutes ago [-]
Impressive, but my poor GPU is melting :)
artursapek 13 minutes ago [-]
ew the bottom is moldy
ramon156 1 hours ago [-]
Imagine if we start designing GPUs around this technology as opposed to vectors. Imagine what voxel engines would look like. Would love a simulated experience or a small scale that theorizes about this.
bozdemir 2 hours ago [-]
this is awesome, I wonder what's under there, looks black, maybe thats where they mounted and rotated the strawberry...
danybittel 2 hours ago [-]
Yes mounting.. and I can't quite reach all the way from below.
35 minutes ago [-]
brazzy 2 hours ago [-]
Lovely! How was the mechanical setup to ensure that all those shots are consistent, and how long did it take?
danybittel 2 hours ago [-]
I posted some pictures.. takes only 20 min!
1 hours ago [-]
timonoko 2 hours ago [-]
What? KIRI Engine makes splats. I always wondered what 3DGS might mean.
Yes. I knoweth what "splats" are: They are splats of fuzzy blobs on the display surface.
ebolyen 26 minutes ago [-]
[dead]
aminekhd 2 hours ago [-]
[dead]
Rendered at 13:09:12 GMT+0000 (Coordinated Universal Time) with Vercel.
What I love about gaussian splats is the way they degrade - instead of a hard cutoff or LoD changing spheres into cubes etc., they get increasingly "dreamy" - the basic idea is still there, just less detailed.
Take for example this scene:
https://superspl.at/scene/e721ea7c
If you navigate closer to the trees, things around you become blurry - as if the very fabric of reality unraveled.
As I scrolled through the website, I was even more impressed with this one though!
https://superspl.at/scene/c67edb74
(I'm ... still not sure what I'm looking at on TFA, and whether or not my browser configuration fails to fully present the site as intended.... OK, if all you're seeing is a blurred image of a strawberry, yes, you'll want to enable a bunch of JS resources. I'm using uMatrix, several hosts must be enabled.)
[1] https://en.wikipedia.org/wiki/Gaussian_splatting
There is no mesh or model. The visual surface of the strawberry could be made up of blobs spaced far apart physically and not where the surface appears to be.
This is why they are called radiance fields, they model the light not the geometry.
Practically the blobs positions/rotations can be constrained to better physically match the geometry of a strawberry.
Otherwise the splat would fall apart as soon as the viewing angle is changed slightly (Which it absolutely does in many examples on supersplat, you cannot really create an out of distribution view with 3GS, it's not magic)
It has a mean, and that mean doesn’t have to lie on the surface, consider the case where the mean is deep inside the strawberry but its spike contributes to the surface appearance (e.g a seed could be represented this way, or it could be represented by a small well-oriented blob on the surface, the optimiser doesn’t care)
I guess the number of samples required to generate a GS is the constraint now, but maybe that will get solved.
https://youtu.be/gXug7Kb3p4I
The filesize of a 3d animated splat is seemingly very small, and the method enables ~arbitrary FPS. But it seems the setup required to record it is still huge and expensive, which limits its usefulness.
Even with that there are some interesting use cases, eg. I'd love to be able to watch concerts this way, and freely move around the stage and crowd from any angle.
For example this bumblebee: https://superspl.at/scene/cf6ac78e
Edit: I completely missed that this was posted by him (:
Focus stacking generally is not perfect process and can lead to artifacts/errors and I'd imagine those can then compound when stacked images are used for 3dgs. Also the image focus actually provides some depth data in itself that could be useful?
My take.. at a macro scale, the dof is usually so small, that it's hard to get a reliably track. So you'd need some sort of way to tell that these stacked photos belong into a series, and then you sort of are doing focus stacking :-) I do think the alignment algorithm could be improved. Maybe the approaches I linked could be used to make a much more robust focus stacking algorithm, that also corrects for 3D geometry. That would be really cool!
You would have to mask out the blurry areas for each image. I guess one could just implement a feature where the optimizer only optimizes gaussians within the sharp distances relative to the camera.
Another way would be some kind of 4d GS where one dimension is the focus distance. But I'd guess the renders would inherently have shallow depth then, which is less useful usually.
I'm wondering if the splat community has decided this paper is valuable -- https://github.com/fraunhoferhhi/Self-Organizing-Gaussians -- looking at all the detail in the strawberry splat made me wonder how small one can get the download, and what the current state of the art is for compression.
Edit. TIL Poe's law
Anyway, very cool splat, fair play
(Can we do a Gaussian Splat of the setup of the photograph for the Gaussian Splat of the Strawberry?)
Yes. I knoweth what "splats" are: They are splats of fuzzy blobs on the display surface.