Research map / Label-efficient and robust learning / Self-supervised and contrastive
Learning Where to Look: Self-supervised Viewpoint Selection for Active Localization using Geometrical Information
Luca Di Giammarino, Boyang Sun, Giorgio Grisetti, Marc Pollefeys, Hermann Blum, Daniel Barath
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 |
|---|---|---|---|
| learn viewpoint score with an MLP on binned landmark reprojections instead of Fisher information field | active viewpoint selection · Habitat-Matterport 3D test, 5m 10deg | localization score (%): 69.6 -> 81.7 | Improved |
| choose viewpoints by uniform binned feature reprojection (no learning) vs Fisher information | baseline · Habitat-Matterport 3D test, 0.25m 2deg | localization score (%): 67.7 -> 71.7 | Improved |
| add learned MLP on top of reprojection-with-bins heuristic | viewpoint selection · Habitat-Matterport 3D test, 5m 10deg | localization score (%): 71.9 -> 81.7 | Improved |
| add depth (z) to 2D landmark projections as input (ablation) | viewpoint MLP · Habitat-Matterport 3D test, 5m 10deg | localization score (%): 76.2 -> 80.6 | Improved |
| use full 3D landmarks instead of depth only (ablation) | viewpoint MLP · Habitat-Matterport 3D test, 0.05m 0.4deg | localization score (%): 74.6 -> 72.9 | Hurt |
| use full 3D landmarks instead of depth only (ablation) | viewpoint MLP · Habitat-Matterport 3D test, 5m 10deg | localization score (%): 80.6 -> 81.7 | Improved |
| use GPU CUDA implementation for viewpoint sampling | sampling pipeline · 648 locations, 24919 landmarks | runtime: 7387 s -> 23 s | Improved |
Closest papers by meaning
- View Consistent Purification for Accurate Cross-View Localization (ICCV 2023 · similarity 0.88)
- Geometry-Aware Learning of Maps for Camera Localization (CVPR 2018 · similarity 0.87)
- Mapping, Localization and Path Planning for Image-Based Navigation Using Visual Features and Map (CVPR 2019 · similarity 0.87)
- LOCATE 3D: Real-World Object Localization via Self-Supervised Learning in 3D (ICML 2025 · similarity 0.87)
- Where Am I and What Will I See: An Auto-Regressive Model for Spatial Localization and View Prediction (ICLR 2025 · similarity 0.87)
- CrowdDriven: A New Challenging Dataset for Outdoor Visual Localization (ICCV 2021 · 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.
