AI Paper Atlas

Research map / 3D vision / 3D generation

Bolt3D: Generating 3D Scenes in Seconds

Stanislaw Szymanowicz, Jason Y. Zhang, Pratul Srinivasan, Ruiqi Gao, Arthur Brussee, Aleksander Holynski, Ricardo Martin-Brualla, Jonathan T. Barron, Philipp Henzler

ICCV 2025 · 3D vision › 3D generation · approach: Score distillation for text-to-3D · 41 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
use generative latent diffusion (Bolt3D) instead of regressionFlash3D · RE10K 1-viewPSNR: 17.40 -> 21.03Improved
use generative approach instead of DepthSplat regressionDepthSplat · DL3DV 4-viewPSNR: 19.48 -> 20.64Improved
use Bolt3D instead of optimization-based CAT3DCAT3D · RE10K 3-viewPSNR: 29.56 -> 27.00Hurt
use Bolt3D instead of optimization-based CAT3D (cost)CAT3D · RE10K 3-viewGPU-min: 77.28 -> 0.25Improved
use Bolt3D instead of LatentSplat (adversarial VAE-GAN)LatentSplat · RE10K 2-viewFID: 2.79 -> 3.66Hurt
train geometry encoder rather than freezing pre-trained one (ablation)Geometry VAE · Pointmap autoencodingRel.: 1.63 -> 0.99Improved
add rec loss re-weighting on distant points (ablation)Geometry VAE · Pointmap autoencodingRel.: 1.68 -> 0.99Improved
add gradient loss (ablation)Geometry VAE · Pointmap autoencodingRel.: 1.16 -> 0.99Improved
use geometry-trained autoencoder instead of pre-trained image autoencoder (ablation)Geometry AE · Real/unbounded pointmapsRel.x100: 17.9 -> 0.670Improved
use ViT decoder instead of conv decoder (ablation)Geometry AE · Pointmaps real datadelta 1.01 / reprojection: 81.4 -> 81.5No clear effect
add cross-attention in Gaussian head (ablation)Gaussian head · Appearance ablationPSNR: 23.80 -> 24.72Improved
add Gaussian head (ablation)Bolt3D · Appearance ablationPSNR: 21.94 -> 24.72Improved

Later papers that compared with this method

1 later accepted paper reports Bolt3D as a baseline in their results tables; 1 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 2 of 3 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