AI Paper Atlas

Research map / 3D vision / Human pose estimation

Object Reprojection Error (ORE): Camera pose benchmarks from lightweight tracking annotations

Xingyu Chen, Weiyao Wang, Hao Tang, Matt Feiszli

NeurIPS 2023 · 3D vision › Human pose estimation · approach: Temporal video pose estimation · 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
replace exhaustive matching with vocabulary tree matching (ablation)COLMAP · EgoStaticORE: 0.066 -> 0.041Improved
replace exhaustive matching with vocabulary tree matching (ablation)COLMAP · EgoStaticruntime: about 1.5x fasterImproved
run on raw fisheye videos without undistortion (ablation)COLMAP-Tuned · EgoStaticORE: 0.041 -> 0.044No clear effect
filter dynamic components with hand-object masks (ablation)COLMAP · EgoStaticORE: 0.066 -> 0.049Improved
remove global backend bundle adjustment (ablation)DROID-SLAM · EgoStaticORE: 0.185 -> 0.214Hurt
remove global backend bundle adjustment (ablation)DROID-SLAM · ScanNetORE: 0.029 -> 0.053Hurt
remove global backend bundle adjustment (ablation)DROID-SLAM · ScanNetATEtrans: 0.147 -> 0.218Hurt
remove global backend bundle adjustment (ablation)DROID-SLAM · ScanNetATErot: 0.135 -> 0.232Hurt
shrink object bounding box for ORE from 1000% to 20% (ablation)ORE metric · ScanNetSpearman correlation with ATErot: 0.384 -> 0.800Improved
shrink object bounding box for ORE from 1000% to 20% (ablation)ORE metric · ScanNetSpearman correlation with ATEtrans: 0.262 -> 0.716Improved
use point tracklets (box size 0%) instead of 20% box (ablation)ORE metric · ScanNetSpearman correlation with ATEtrans: 0.716 -> 0.720No clear effect
move from ScanNet to egocentric EgoStatic videosDROID-SLAM · EgoStaticORE: 0.029 -> 0.185Hurt

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