Trendoday

SplatAnything: Gaussian Spacing Normalization for Feed-forward 3DGS across Arbitrary Scenes

We identify a key limitation of existing feed-forward 3DGS methods: Gaussian attributes are typically predicted in a radial-distance-normalized space that canonicalizes a global radial statistic of th

  • Upvotes38
–Rank now
#1Peak rank
2Days on list
414Score · 7d
Nov 9, 2026First seen
Rank · 7 days

About this paper

SplatAnything applies Gaussian Spacing Normalization to canonicalize scene-level Gaussian spacing statistics, improving feed-forward 3DGS reconstruction across varying scene scales and viewpoints.

We identify a key limitation of existing feed-forward 3DGS methods: Gaussian attributes are typically predicted in a radial-distance-normalized space that canonicalizes a global radial statistic of the reconstructed geometry, while the spacing between neighboring Gaussian centers can still vary substantially across scene scales and camera configurations. This mismatch induces systematic shifts in Gaussian-scale targets, leading to ambiguous supervision. To address this, we propose SplatAnything with Gaussian Spacing Normalization (GSNorm), which canonicalizes a robust scene-level statistic of Gaussian spacing to provide a more consistent spatial reference for Gaussian prediction. To…

feed-forward 3DGS methodsGaussian attributesradial-distance-normalized spaceglobal radial statisticGaussian centersGaussian-scale targetsambiguous supervisionSplatAnythingGaussian Spacing Normalization (GSNorm)scene-level statistic
Upvotes
38
Authors
10
Published
Nov 5, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

SplatAnything: Gaussian Spacing Normalization for Feed-forward 3DGS across Arbitrary Scenes was first seen on Hugging Face · Daily papers on Nov 9, 2026. Since then it has appeared in 17 checks on 2 different days. Its best position was #1 of 30, reached on Nov 10, 05:28 UTC. It dropped off the list after Nov 10, 15:28 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 07:2815:2823:2807:2815:28 Nov 9, 07:28 UTC · #9Nov 9, 09:28 UTC · #6Nov 9, 11:28 UTC · #6Nov 9, 13:28 UTC · #6Nov 9, 15:28 UTC · #4Nov 9, 17:28 UTC · #3Nov 9, 19:28 UTC · #3Nov 9, 21:28 UTC · #3Nov 9, 23:28 UTC · #3Nov 10, 01:28 UTC · #2Nov 10, 03:28 UTC · #2Nov 10, 05:28 UTC · #1Nov 10, 07:28 UTC · #2Nov 10, 09:28 UTC · #2Nov 10, 11:28 UTC · #2Nov 10, 13:28 UTC · #2Nov 10, 15:28 UTC · #2

Last 14 days

Best rank per day (UTC).

Nov 3today

Metric history

Platform metric as reported by Hugging Face.

0224467 07:2815:2823:2807:2815:28 Nov 9, 07:28 UTC · 38Nov 9, 09:28 UTC · 54Nov 9, 11:28 UTC · 57Nov 9, 13:28 UTC · 59Nov 9, 15:28 UTC · 59Nov 9, 17:28 UTC · 59Nov 9, 19:28 UTC · 59Nov 9, 21:28 UTC · 60Nov 9, 23:28 UTC · 60Nov 10, 01:28 UTC · 60Nov 10, 03:28 UTC · 64Nov 10, 05:28 UTC · 64Nov 10, 07:28 UTC · 64Nov 10, 09:28 UTC · 66Nov 10, 11:28 UTC · 67Nov 10, 13:28 UTC · 67Nov 10, 15:28 UTC · 67

Questions

Is SplatAnything: Gaussian Spacing Normalization for Feed-forward 3DGS across Arbitrary Scenes trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 10, 15:28 UTC, after 2 days of trending.

What is the highest rank SplatAnything: Gaussian Spacing Normalization for Feed-forward 3DGS across Arbitrary Scenes reached?

#1 on Hugging Face · Daily papers, on Nov 10, 05:28 UTC. Trendoday checks this list every 120 minutes.

How long has SplatAnything: Gaussian Spacing Normalization for Feed-forward 3DGS across Arbitrary Scenes been trending?

It was first seen on Nov 9, 2026 and has been on the list on 2 days (17 checks) since.