AI Paper Atlas

Research map / 3D vision / 3D generation

latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction

Christopher Wewer, Kevin Raj, Eddy Ilg, Bernt Schiele, Jan E. Lenssen

ECCV 2024 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 90 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add generative variational formulation and GAN loss (vs deterministic) (ablation)latentSplat · CO3D hydrants extrapolationFID: 69.33 -> 48.88Improved
add variational formulation and GAN loss (vs deterministic) (ablation)latentSplat · CO3D hydrants extrapolationPSNR: 16.50 -> 15.61Hurt
add multi-scale latent skip connections to decoder (ablation)latentSplat · CO3D hydrants extrapolationFID: 50.31 -> 48.88No clear effect
add RGB skip input to decoder (ablation)latentSplat · CO3D hydrants extrapolationFID: 48.48 -> 48.88No clear effect
replace pixelSplat's regression with latentSplat variational Gaussians and decoderpixelSplat · RealEstate10k extrapolationFID: 5.78 -> 2.79Improved
replace pixelSplat's regression with latentSplat on CO3D hydrantspixelSplat · CO3D hydrants interpolationPSNR: 18.05 -> 18.01No clear effect
use latentSplat output images for mesh reconstruction instead of pixelSplat2D Gaussian Splatting mesh · CO3D hydrantsChamfer distance (x10^-8): 1.815 -> 1.535Improved
use generative decoder instead of GeNVS diffusion samplingnovel view synthesis · efficiencyRender time (s): 6.423 -> 0.003Improved
add generative decoder on top of Gaussian splattingpixelSplat · render timeRender time (s): 0.002 -> 0.003Hurt

Later papers that compared with this method

6 later accepted papers report LatentSplat as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy