AI Paper Atlas

Research map / 3D vision / 3D generation

LART: Neural Correspondence Learning with Latent Regularization Transformer for 3D Motion Transfer

Haoyu Chen, Hao Tang, Radu Timofte, Luc Van Gool, Guoying Zhao

NeurIPS 2023 · 3D vision › 3D generation · approach: Learning physics from video · 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
add latent metric regularization (LMR) to LART (ablation)LART · DFAUST single subjectPMD (x10^-4): 1.90 -> 0.32Improved
add latent metric regularization (LMR) to LART (ablation)LART · DFAUST single subjecttraining time (h): 20 h -> 8 hImproved
feed multi-frame driving sequence instead of single frame (ablation)LART · DFAUST single subjectPMD (x10^-4): 2.77 -> 0.32Improved
feed multi-frame driving sequence instead of single frame (ablation)LART · DFAUST single subjecttraining time (h): 25 h -> 8 hImproved
replace NPT with LART (Transformer with multi-frame input and LMR)NPT · DFAUST single subjectPMD (x10^-4): 2.82 -> 0.32Improved
single-frame LART versus NPT baseline (ablation)LART single frame · DFAUST single subjectPMD (x10^-4): 2.82 -> 2.77No clear effect
replace 3D-CoreNet with LART3D-CoreNet · DFAUST seen motionsPMD (x10^-4): 1.25 -> 0.41Improved
replace 3D-CoreNet with LART3D-CoreNet · DFAUST unseen motionsPMD (x10^-4): 3.05 -> 1.80Improved
replace NPT with LARTNPT · DFAUST unseen motionsPMD (x10^-4): 5.31 -> 1.80Improved
evaluate on unseen instead of seen driving motionsLART · DFAUSTPMD (x10^-4): 0.41 -> 1.80Hurt
use multi-frame processing instead of frame-by-frame deformation on unseen motionsLART vs NPT and 3D-CoreNet · AMASS long sequencesqualitative robustnessImproved
add NPT motion-prior variant (NPT-MP) over NPTNPT · DFAUST seen motionsPMD (x10^-4): 4.37 -> 4.51Hurt

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