AI Paper Atlas

Research map / 3D vision / Depth and stereo

Discrete Laplace Operator Estimation for Dynamic 3D Reconstruction

Xiangyu Xu, Enrique Dunn

ICCV 2019 · 3D vision › Depth and stereo · approach: Feed-forward 3D reconstruction · 10 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 DTW arc distance instead of Euclidean distance for sequencing (SR, nonlinear motion, initial structure) (ablation)DLOE sequencing · Motion capture (nonlinear motion)Kendall rank correlation: 0.9807 -> 0.9570Hurt
use DTW arc distance instead of Euclidean distance for sequencing (MDS, optimized structure, nonlinear motion) (ablation)DLOE sequencing · Motion capture (nonlinear motion)Kendall rank correlation: 0.7044 -> 1Improved
use DTW arc distance instead of Euclidean distance for sequencing (SR, initial structure, repeating motion) (ablation)DLOE sequencing · Motion capture (repeating motion)Kendall rank correlation: 0.6754 -> 0.9711Improved
use DTW arc distance instead of Euclidean distance for sequencing (SHP, optimized structure, repeating motion) (ablation)DLOE sequencing · Motion capture (repeating motion)Kendall rank correlation: 0.7912 -> 0.7934No clear effect
sequence from optimized structure instead of initial structure (SHP, Euclidean, nonlinear motion) (ablation)DLOE sequencing · Motion capture (nonlinear motion)Kendall rank correlation: 0.4368 -> 0.9996Improved
sequence from optimized structure instead of initial structure (MDS, Euclidean, nonlinear motion) (ablation)DLOE sequencing · Motion capture (nonlinear motion)Kendall rank correlation: 0.7614 -> 0.7044Hurt
sequence from optimized structure instead of initial structure (SHP, arc distance, repeating motion) (ablation)DLOE sequencing · Motion capture (repeating motion)Kendall rank correlation: 0.3947 -> 0.7934Improved
include all geometric terms (smoothness, locality, reconstructability) in the cost function (ablation)DLOE · Motion capture, moderate-to-high 2D noisereconstruction errorImproved
increase 2D observation noise from 1 to 5 pixels stdDLOE · Motion capture3D reconstruction errorHurt
use DLOE without sequencing instead of SEDL at lower frame ratesDLOE · Motion capture (30, 15, 7.5 Hz)3D reconstruction errorImproved
use DLOE instead of SEDL and TB under 10-50% missing pointsDLOE · Motion capture3D reconstruction errorImproved
drop 10-50% of frames (non-uniform density), DLOE vs other methodsDLOE · Motion capture3D reconstruction errorImproved

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