AI Paper Atlas

Research map / 3D vision / 3D generation

Sparse auto-regressive modeling for scene generation from multi-view images

Thomas Lucas, Maxime Pietrantoni, Wonjune Cho, Bardienus Duisterhof, Philippe Weinzaepfel, Vincent Leroy, Jerome Revaud

ECCV 2026 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 0 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
generative sparse voxel latent AR model vs best baselineSPAR3S · 3DFront 2-viewFID: 110 -> 59 (vs DepthSplat)Improved
generative sparse voxel latent AR model vs LatentSplatSPAR3S · RealEstate10k 2-viewPSNR: 16.09 -> 16.72Improved
add bincount-based guided attention (ablation)SPAR3S encoder · 3DFronttraining speed and PSNRImproved
increase number of Gaussians per voxel (ablation)SPAR3S decoder · 3DFrontPSNRImproved
inject about 25% false-positive voxels at training upsampling (ablation)SPAR3S encoder · 3DFrontPSNRImproved
KL coefficient above 0.1 or latent dim below 32 (ablation)SPAR3S encoder · 3DFrontPSNRHurt
replace diffusion head with linear head (ablation)SPAR3S AR model · 3DFrontnovel view qualityHurt
add 3D relative positional encoding (ablation)SPAR3S AR model · 3DFrontnovel view qualityImproved
use BFS region-growing ordering instead of random order (ablation)SPAR3S inference · 3DFrontnovel view qualityImproved
add occupancy refinement (coarse-to-fine) (ablation)SPAR3S · 3DFrontnovel view qualityImproved
increase number of conditioning views 1 to 12SPAR3S · 3DFrontPSNR / LPIPS: PSNR 11.73 -> 19.62 at threshold 0.1 (1 -> 12 views)Improved
use ground truth occupancy (ablation)SPAR3S · 3DFrontnovel view qualityImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 5 of 5 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