AI Paper Atlas

Research map / 3D vision / Face and shape

Bootstrapping Articulated 3D Reconstruction from 2D Image Collections

Jakub Zadrozny, Oisin Mac Aodha, Hakan Bilen

ECCV 2026 · 3D vision › Face and shape · approach: Articulated and non-rigid shape learning · 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
use heavy-tailed Student's t camera sampling for the canonical initialization (ablation)canonical initialization · Horse-Mixedmean CD (cm): 10.04 -> 7.24Improved
add iterative refinement from image collection without fitting regularization (ablation)BAT3R · Horse-Mixedmean CD (cm): 7.24 -> 5.59Improved
add geometric regularization in mesh fitting (ablation)BAT3R · Horse-Mixedmean CD (cm): 5.59 -> 4.85Improved
add viewpoint augmentation when rendering fitted meshes (ablation)BAT3R · Horse-Mixedmean CD (cm): 4.85 -> 3.86Improved
render fitted meshes into synthetic images instead of pairing raw images with estimated 3D (ablation)BAT3R · Horse-Mixedmean CD (cm): 6.96 -> 3.86Improved
fit realistic articulations from images instead of random pose sampling (ablation)BAT3R · Horse-Mixedmean CD (cm): 5.22 -> 3.86Improved
train on real in-the-wild horse images instead of DualPM synthetic images (ablation)BAT3R · Horse-RobustRMS CD (cm): 5.10 -> 6.01Hurt
refine from canonical initialization using synthetic imagesBAT3R · Horse-RobustRMS CD (cm): 7.84 -> 5.10Improved
train on real images vs canonical initializationBAT3R · Horse-RobustRMS CD (cm): 7.84 -> 6.01Improved
compare with fully supervised DualPMBAT3R (synt) vs DualPM · Animodel-Points HorseRMS CD (cm): 4.30 vs 5.65Hurt
number of refinement iterations (ablation)BAT3R · Horse-Mixedmean CD / outlier ratio: first iteration large jump, then plateauMixed

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