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

Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

svraster.github.io

Post reply on HN
Print view View original

11–20 of 20 posts

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#11

Post by chpatrick »

bondarchuk wrote: post #10
loxias wrote: post #7

I look forward to reading this in closer detail, but it looks like they solve an inverse problem to recover a ground truth set of voxels (from a large set of 2d images with known camera parameters), which is underconstrained. Neat to me that it works w/o using dense optical flow to recover the structure -- I wouldn't have thought that would converge. Love this a whole heck of a lot more than NeRF, or any other "lol l…

> Love this a whole heck of a lot more than NeRF, or any other "lol lets just throw a huge network at it" approach. Well yes, but that's what gaussian splatting also was. The question is: are their claims to be so much better than gsplat accurate?

There's no neural net with gaussian splatting, it's a fancy pointcloud that's optimized with ML techniques.

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#12

Post by bondarchuk »

Funny, it almost sounds like a straight efficiency improvement of Plenoxels (the direct predecessor of gaussian splatting), which would mean gaussian splatting was something of a a red herring/sidetrack. Though I'm not sure atm where the great performance gain is. Definitely interesting.

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#13

Post by bondarchuk »

chpatrick wrote: post #11
bondarchuk wrote: post #10

Earlier quoted context omitted.

> Love this a whole heck of a lot more than NeRF, or any other "lol lets just throw a huge network at it" approach. Well yes, but that's what gaussian splatting also was. The question is: are their claims to be so much better than gsplat accurate?

There's no neural net with gaussian splatting, it's a fancy pointcloud that's optimized with ML techniques.

I know, that's the point.

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#14

Post by chpatrick »

bondarchuk wrote: post #13
chpatrick wrote: post #11

Earlier quoted context omitted.

There's no neural net with gaussian splatting, it's a fancy pointcloud that's optimized with ML techniques.

I know, that's the point.

Mea culpa, I misunderstood.

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#15

Post by ibrarmalik »

bondarchuk wrote: post #12

Funny, it almost sounds like a straight efficiency improvement of Plenoxels (the direct predecessor of gaussian splatting), which would mean gaussian splatting was something of a a red herring/sidetrack. Though I'm not sure atm where the great performance gain is. Definitely interesting.

How is plenoxels a direct predecessor of gaussian splatting?

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#17

Post by yyeboah »

ibrarmalik wrote: post #15
bondarchuk wrote: post #12

Funny, it almost sounds like a straight efficiency improvement of Plenoxels (the direct predecessor of gaussian splatting), which would mean gaussian splatting was something of a a red herring/sidetrack. Though I'm not sure atm where the great performance gain is. Definitely interesting.

How is plenoxels a direct predecessor of gaussian splatting?

They both emerged out of the pursuit of a more efficient solution for addressing the inefficiencies in NeRF, which was mainly due to expensive ray marching and MLP calls. Before the emergence of Gaussian splatting, grids, such as plenoxels were all the rage. Of course, Gaussian splatting here refers to the paper, “3D Gaussian Splatting for Real-Time Radiance Field Rendering”

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#18

Post by yorwba »

HexDecOctBin wrote: post #16

Why is this called rendering, when it would be more accurate to call it reverse-rendering (unless "rendering" means any kind of transformation of visual-adjacent data)?

The reverse-rendering is not real-time, but takes several minutes. Only rendering new viewpoints from the resulting sparse voxel representation runs at high enough framerates.

Re: Sparse Voxels Rasterization: Real-Time High-Fidelity Radiance Field Rendering

#19

Post by markisus »

This is basically Gaussian splat using cubes instead of Gaussians. The cube centers and sizes choices are discrete and non overlapping, hence the name “sparse voxel”. The qualitative results and rendering speeds are similar to Gaussian splat, and it’s sometimes better or worse depending on the scene.
Post reply on HN