AI Paper Atlas

Research map / 3D vision / Face and shape

DiffSurf: A Transformer-based Diffusion Model for Generating and Reconstructing 3D Surfaces in Pose

Yusuke Yoshiyasu, Leyuan Sun

ECCV 2024 · 3D vision › Face and shape · approach: Full-body avatar reconstruction · 2 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
replace transformer layers with MLP layers in the diffusion network (ablation)DiffSurf · SURREAL unconditional generation1-NNA: 55.6 -> 92.6Hurt
add long skip connections to the diffusion transformer (ablation)DiffSurf · SURREAL unconditional generation1-NNA: 54.4 -> 55.6Hurt
use 3D template position embedding instead of learned (ablation)DiffSurf · SURREAL unconditional generation1-NNA: 55.6 -> 54.8No clear effect
add time embedding instead of using a time token (ablation)DiffSurf · SURREAL unconditional generation1-NNA: 55.6 -> 55.4No clear effect
increase DDIM sampling steps from 1 to 10-20 (ablation)DiffSurf human mesh recovery · 3DPWPA-MPJPE: 143.9 -> 54.3 (10 steps)Improved
increase sampling steps from 10 to 100 (ablation)DiffSurf human mesh recovery · 3DPWPA-MPJPE / fps: 54.3 -> 54.7; fps 21.2 -> 3.38Hurt
increase CFG scale from 0 to 1.0 (ablation)DiffSurf human mesh recovery · 3DPWPA-MPJPE: 52.6 -> 54.3Hurt
increase CFG scale from 0 to 0.5 (ablation)DiffSurf human mesh recovery · Human3.6MPA-MPJPE: 37.9 -> 36.1Improved
add SDS-based fitting using DiffSurf prior to HMR-EFT with GT 2D keypointsHMR-EFT · 3DPW fittingPA-MPJPE: 54.3 -> 52.1Improved
add SDS-based fitting using DiffSurf prior to METROMETRO · 3DPW fittingPA-MPJPE: 63.0 -> 54.2Improved
add L_consist loss in control point deformation (ablation)DiffSurf control point deformation · qualitative control point deformationfit and distortionMixed
add edge and Laplacian SDS losses in control point deformation (ablation)DiffSurf control point deformation · qualitative control point deformationdistortionImproved

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