Research map / 3D vision / Depth and stereo
LINK: Learning to Localize from Known to Unknown Scenes
Minghang Zhu, Kaibo Jin, Zhijing Wang, Ziwei Shi, Wen Li, Sheng Ao, Cheng Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes LINK, a confidence-guided LiDAR relocalization framework based on scene coordinate regression that handles test trajectories traversing route segments unseen in training.
Problem
LiDAR relocalization with scene coordinate regression degrades sharply when test trajectories pass through route segments absent from training, a common issue in long-range autonomous driving where full route coverage is rarely possible.
Approach
A confidence-guided framework that predicts an absolute pose hypothesis together with scene features and uses confidence to handle partial scene coverage; the abstract is truncated so the exact mechanism is not fully stated.
Closest papers by meaning
- LightLoc: Learning Outdoor LiDAR Localization at Light Speed (CVPR 2025 · similarity 0.90)
- LiSA: LiDAR Localization with Semantic Awareness (CVPR 2024 · similarity 0.89)
- DiffLoc: Diffusion Model for Outdoor LiDAR Localization (CVPR 2024 · similarity 0.88)
- ICP-Flow: LiDAR Scene Flow Estimation with ICP (CVPR 2024 · similarity 0.87)
- Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis (NeurIPS 2026 · similarity 0.87)
- SOAR: Regression-based LiDAR Relocalization for UAVs (NeurIPS 2026 · similarity 0.87)
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.
