AI Paper Atlas

Research map / 3D vision / Face and shape

2D-LFM: Lifting Foundation Model without 3D Supervision

Mosam Dabhi, Irhas Gill, László A. Jeni, Simon Lucey

CVPR 2026 · 3D vision › Face and shape · approach: Single-view 3D reconstruction

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
inject positional encoding at every attention layer instead of only at the input (ablation)2D-only transformer lifter · Pascal3D+MPJPE (mm): 100.3 -> 8.1Improved
inject positional encoding at every attention layer instead of only at the input (ablation)2D-only transformer lifter · Human3.6MMPJPE (mm): 63.4 -> 33.1Improved
inject PE only at first and last layers (ablation)2D-only transformer lifter · Pascal3D+MPJPE (mm): 100.3 -> 92.1No clear effect
inject PE at every 2nd layer rather than every layer (ablation)2DLFM · Pascal3D+MPJPE (mm): 8.1 -> 26.1Hurt
use analytical RFF/Fourier PE instead of ViT-style sinusoidal PE (ablation)2DLFM · Pascal3D+MPJPE (mm): 92.3 -> 8.1Improved
swap Fourier PE for graph Laplacian PE (both per-layer) (ablation)2DLFM · Pascal3D+ and Human3.6MMPJPE (mm): RFF 11.2/38.1 vs Laplacian 9.3/35.8No clear effect
increase transformer depth with per-layer PE (ablation)2DLFM · Pascal3D+MPJPE (mm): 15.3 (4 layers) -> 9.3 (12 layers) -> 8.1 (24 layers)Improved
train a single combined multi-category model instead of per-category models (ablation)2DLFM · Pascal3D+ bottleMPJPE (mm): 100 -> 7.2Improved
train a single combined multi-category model instead of per-category models (ablation)2DLFM · Pascal3D+ carMPJPE (mm): 8.9 -> 6.7Improved
learn from 2D supervision only vs 3D-supervised 3D-LFM2DLFM · Human3.6MMPJPE (mm): 46.3 -> 30.9Improved
learn from 2D supervision only vs 3D-supervised 3D-LFM2DLFM · Pascal3D+MPJPE (mm): 5.2 -> 8.1Hurt
use landmark lifting instead of back-projecting landmarks through VGGT depth2DLFM · Human3.6MMPJPE (mm): 107.8 -> 30.9Improved

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