Research map / Point clouds / Point cloud classification
Yes, we CANN: Constrained Approximate Nearest Neighbors for Local Feature-Based Visual Localization
Dror Aiger, Andre Araujo, Simon Lynen
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 |
|---|---|---|---|
| use CANN (joint appearance and geometry constrained NN with local features only) instead of global-feature retrieval for localization | CANN · four localization benchmarks (Baidu, Gangnam, RobotCar, Aachen) | retrieval EWB and SfM localization | Improved |
| use CANN-RG instead of ASMK for local feature retrieval (index time) | HOW features · Baidu-Mall | index time: 253.34 -> 9.08 | Improved |
| use CANN-RG instead of ASMK (query time) | HOW features · RobotCar | query time per image (s): 0.53 -> 0.04 | Improved |
| use CANN-RS instead of ASMK (query time) | HOW features · RobotCar | query time per image (s): 0.53 -> 5.29 | Hurt |
| use CANN-RG (approximate) instead of CANN-RS (ablation) | CANN · Baidu-Mall | simplified EWB top-1: 12.47 -> 12.12 | No clear effect |
| use CANN-RG instead of CANN-RS (ablation) | CANN · indexing time (Aachen) | index time: 11.19 -> 244.09 | Hurt |
| add CANN-HOW weighted with DELG global features | DELG · ROxford Hard | mAP: 55.6 -> 64.2 | Improved |
| weighted DELG+CANN-FIRE vs best global-only (CVNet-Global) | ROxford Hard · ROxford Hard | mAP: 63.1 -> 62.3 | Hurt |
Closest papers by meaning
- Investigating the Role of Image Retrieval for Visual Localization (IJCV 2022 · similarity 0.91)
- Mapping, Localization and Path Planning for Image-Based Navigation Using Visual Features and Map (CVPR 2019 · similarity 0.89)
- Towards Visual Localization Interoperability: Cross-Feature for Collaborative Visual Localization and Mapping (ICCV 2025 · similarity 0.89)
- Are Large-Scale 3D Models Really Necessary for Accurate Visual Localization? (CVPR 2017 · similarity 0.89)
- RELOCATE: A Simple Training-Free Baseline for Visual Query Localization Using Region-Based Representations (CVPR 2025 · similarity 0.88)
- R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization (CVPR 2025 · 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.
