AI Paper Atlas

Research map / 3D vision / Human pose estimation

Learned Vertex Descent: A New Direction for 3D Human Model Fitting

Enric Corona, Gerard Pons-Moll, Guillem Alenyà, Francesc Moreno-Noguer

ECCV 2022 · 3D vision › Human pose estimation · approach: Human shape and pose regression · 44 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 IP-Net with LVD for SMPL registrationSMPL registration · Body registration (SMPL)Joint error [cm]: 3.60 -> 2.53Improved
replace IP-Net with LVD for SMPL registrationSMPL registration · Body registration (SMPL)Vertex error [cm]: 5.03 -> 3.00Improved
replace no-correspondence fitting with LVD initializationSMPL registration · Body registration (SMPL)Joint error [cm]: 16.6 -> 2.53Improved
add iterative LVD optimization on top of the forward pass (ablation)LVD · Body registration (SMPL)Joint error [cm]: 5.89 -> 2.53Improved
add iterative LVD optimization on top of the forward pass (ablation)LVD · Body registration (SMPL)Vertex error [cm]: 6.27 -> 3.00Improved
use LVD instead of Sengupta on BUFFsingle-view SMPL · BUFF single-view SMPLV2V avg [mm]: 43.05 -> 39.08Improved
use LVD instead of FrankMocap on BUFFsingle-view SMPL · BUFF single-view SMPLV2S avg [mm]: 38.66 -> 37.87Improved
use LVD instead of Sengupta for back view (180 deg)single-view SMPL · BUFF single-view SMPLV2V at 180 deg [mm]: 40.11 -> 54.55Hurt
use LVD instead of IP-Net for MANO registrationhand registration · Hand registration (MANO)Joint error [cm]: 1.44 -> .76Improved
use LVD instead of IP-Net for MANO registrationhand registration · Hand registration (MANO)Scan-to-recons V2S [mm]: 4.08 -> 3.33Improved
use deeper MLP or positional encoding (ablation)LVD · LVD architectureperformanceNo clear effect
refine SMPL registration to SMPL+DLVD · Body registrationRecons-to-scan V2V [mm]: 8.16 -> 1.14Improved

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