AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

RTMap: Real-Time Recursive Mapping with Change Detection and Localization

Yuheng Du, Sheng Yang, Lingxuan Wang, Zhenghua Hou, Chengying Cai, Zhitao Tan, Mingxia Chen, Shi-Sheng Huang, Qiang Li

ICCV 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 9 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
add probabilistic vertex density U (ablation)RTMap cycle 3 · TbV straightmAP avg: 50.7 -> 57.6Improved
add probabilistic vertex density U (ablation)RTMap cycle 2 · TbV straightmAP avg: 40.2 -> 45.6Improved
add probabilistic vertex density U (ablation)RTMap cycle 3 · TbV turningmAP avg: 46.5 -> 55.4Improved
add probabilistic vertex density U (ablation)RTMap cycle 2 · TbV turningmAP avg: 41.9 -> 45.5Improved
add more crowdsourcing cycles (2 to 3)RTMap · TbV straightmAP avg: 45.6 -> 57.6Improved
use RTMap crowdsourcing instead of MapTracker averagingMapTracker · TbV turningmAP avg: 28.0 -> 55.4Improved
add change detection by separating Q_map from Q_prior (ablation)RTMap · TbV localizationlateral error mean (m): 0.163 -> 0.125Improved
add change detection by separating Q_map from Q_prior (ablation)RTMap · TbV localizationyaw error mean (deg): 0.332 -> 0.317Improved
use optimization-based pose T^R instead of end-to-end T^E (ablation)RTMap · nuScenes localizationyaw error mean (deg): 0.521 -> 0.368Improved
use optimization-based pose T^R instead of end-to-end T^E (ablation)RTMap · nuScenes localizationlateral error mean (m): 0.142 -> 0.121Improved
add localization loss L_pose (ablation)RTMap T^R · nuScenes localizationlongitudinal error 90th (m): 1.510 -> 1.436Improved
add vertex regression loss L_pts (ablation)RTMap T^R · nuScenes localizationlateral error mean (m): 0.122 -> 0.118Improved

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