AI Paper Atlas

Research map / 3D vision / Depth and stereo

Latent Fusion: Decoding Consolidated 3D Geometry from Feed-forward Geometry Transformer Latents

Laura Fink, Linus Franke, George Kopanas, Marc Stamminger, Peter Hedman

ECCV 2026 · 3D vision › Depth and stereo · approach: Cost-volume multi-view stereo

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
Regress SDF directly from transformer latents instead of fusing per-view predictions (TSDF)VGGT_ft · Objaverse 8 viewsF-score at 0.5 eps: 0.43 -> 0.51Improved
Regress SDF directly instead of TSDF fusionVGGT_ft · Objaverse 8 viewsSDF MAE: 0.023 -> 0.004Improved
Regress SDF directly instead of TSDF fusionVGGT_ft · Objaverse 8 viewsEMD: 0.014 -> 0.019Hurt
Regress SDF directly instead of TSDF fusionVGGT_ft · Objaverse 8 viewsChamfer distance: 0.022 -> 0.021No clear effect
Replace Poisson fusion with latent SDF extractionVGGT_ft · Objaverse 8 viewsF-score at eps: 0.70 -> 0.83Improved
Replace per-view fusion with joint latent extraction at 24 views (trained on up to 8)Fus3D · Objaverse varying viewsF-score / ChamferImproved
Use Fus3D over UFORecon with VGGT poses (unfavorable views)UFORecon · DTU unfavorableChamfer distance: 4.942 -> 3.525Improved
Use Fus3D over VolRecon with VGGT poses (favorable views)VolRecon · DTU favorableChamfer distance: 3.678 -> 2.432Improved
Compare Fus3D with UFORecon given COLMAP poses (unfavorable)UFORecon · DTU unfavorableChamfer distance: 3.275 vs 3.525Mixed
Use learned canonical embedding instead of projected voxel positions (variable query volume) (ablation)Fus3D+ · Objaverse variable queryChamfer distance: 0.021 -> 0.018Improved
Use learned canonical embedding (default query volume) (ablation)Fus3D+ · Objaverse default queryF-score at 0.5 eps: 0.466 -> 0.467No clear effect
Add near-surface SDF supervision term (ablation)Fus3D · Objaverseisosurface accuracyImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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