AI Paper Atlas

Research map / 3D vision / Depth and stereo

Elaborate Monocular Point and Line SLAM With Robust Initialization

Sang Jun Lee, Sung Soo Hwang

ICCV 2019 · 3D vision › Depth and stereo · approach: Scene coordinate regression for localization · 36 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 rank-1 factorization initialization instead of conventional ORB-SLAM initializationORB-SLAM point SLAM · TUM RGB-D f1_xyzATE RMSE (cm): 1.2514 -> 1.0178Improved
use proposed initialization instead of conventionalORB-SLAM point SLAM · TUM RGB-D f2_xyzATE RMSE (cm): 0.3655 -> 0.3062Improved
use proposed initialization instead of conventionalORB-SLAM point SLAM · TUM RGB-D f2_deskRPE (cm): 7.2342 -> 6.7241Improved
use proposed initialization instead of conventionalORB-SLAM point SLAM · TUM RGB-D f3_sit_staticATE RMSE (cm): 0.3783 -> 0.5934Hurt
use proposed initialization instead of conventionalORB-SLAM point SLAM · TUM RGB-D f1_floorATE RMSE (cm): 3.4003 -> 3.4938No clear effect
use proposed initialization where conventional failsORB-SLAM point SLAM · TUM RGB-D f3_nstr_tex_farATE RMSE (cm): ambiguity -> 3.6836Improved
use Plucker line representation (RIPPL-SLAM) instead of endpoints (EPL-SLAM)point-line SLAM · TUM RGB-D f1_xyzATE RMSE (cm): 1.1873 -> 0.9399Improved
use RIPPL-SLAM instead of PPL-SLAMpoint-line SLAM · TUM RGB-D f1_floorATE RMSE (cm): 4.7881 -> 3.4988Improved
use RIPPL-SLAM instead of EPL-SLAMpoint-line SLAM · TUM RGB-D f2_deskATE RMSE (cm): 5.3653 -> 5.5371Hurt
use RIPPL-SLAM instead of PPL-SLAMpoint-line SLAM · TUM RGB-D f2_xyzATE RMSE (cm): 0.3436 -> 0.3907Hurt
use RIPPL-SLAM instead of EPL-SLAMpoint-line SLAM · TUM RGB-D f3_nstr_tex_farATE RMSE (cm): ambiguity -> 1.5060Improved
optimize 4-parameter orthonormal line instead of 6-parameter endpointslocal BA · speedlocal BA time: about 1.5 times fasterImproved

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