Research map / Vision-language and multimodal / Human and embodied modeling
TextPlace: Visual Place Recognition and Topological Localization Through Reading Scene Texts
Ziyang Hong, Yvan Petillot, David Lane, Yishu Miao, Sen Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add a 10 m position prior to NetVLAD | NetVLAD · Street 1 day-night, recall 0.8 | precision: 0.76 -> 0.95 | Improved |
| enlarge the NetVLAD position-prior search space from 10 m to 30 m (ablation) | NetVLAD · Street 1 day-night, recall 0.9 | precision: 0.93 -> 0.83 | Hurt |
| enlarge the FAB-MAP position-prior search space from 10 m to 30 m (ablation) | FAB-MAP · Street 1 day-night, recall 0.9 | precision: 0.63 -> 0.58 | Hurt |
| add VO tracking, topological graph and adaptive search window (versus global search) (ablation) | TextPlace · Street 1 day-night, recall 0.9 | precision: 0.84 -> 0.91 | Improved |
| add VO tracking, topological graph and adaptive search window (versus global search) (ablation) | TextPlace · Street 1 day-night, recall 0.8 | precision: 0.92 -> 0.96 | Improved |
| replace NetVLAD with TextPlace under global search | NetVLAD (global) · Street 1 day-night, recall 0.9 | precision: 0.74 -> 0.84 | Improved |
| increase the search window size (ablation) | TextPlace · V4RL sequence 2 | precision | Improved |
| use denser map nodes (smaller displacement between nodes) (ablation) | TextPlace · V4RL sequence 2 | precision | No clear effect |
| use a larger search window to offset the Levenshtein distance threshold (ablation) | TextPlace · V4RL sequence 2 | precision | Improved |
| raise the Levenshtein threshold to reach recall above 0.9 (ablation) | TextPlace · Mall | precision | Hurt |
Closest papers by meaning
- LaVPR: Benchmarking Language and Vision for Place Recognition (ECCV 2026 · similarity 0.87)
- SAGE: Spatial-visual Adaptive Graph Exploration for Efficient Visual Place Recognition (ICLR 2026 · similarity 0.86)
- Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition (AAAI 2022 · similarity 0.86)
- Text2Loc: 3D Point Cloud Localization from Natural Language (CVPR 2024 · similarity 0.86)
- Textual Supervision Enhances Geospatial Representations in Vision-Language Models (ICML 2026 · similarity 0.85)
- Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition (AAAI 2025 · similarity 0.84)
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.
