Research map / 3D vision / Depth and stereo
Geolocation Estimation of Photos using a Hierarchical Model and Scene Classification
Eric Muller-Budack, Kader Pustu-Iren, Ralph Ewerth
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 scene-specific Individual Scene Networks (S3) to hierarchical multi-partitioning model (ablation) | base (M, f*) · Im2GPS3k test | accuracy within 1 km: 9.7 -> 10.5 | Improved |
| add scene-specific Individual Scene Networks (S3) to hierarchical multi-partitioning model (ablation) | base (M, f*) · Im2GPS test | accuracy within 1 km: 15.2 -> 16.9 | Improved |
| use hierarchical classification (f*) instead of standard finest-cell prediction (ablation) | base (M, m) · Im2GPS3k test | accuracy within 25 km: 25.5 -> 27.0 | Improved |
| train with multiple partitionings instead of single partitioning (ablation) | base (L, m) · Im2GPS test | accuracy within 25 km: 36.3 -> 35.0 | No clear effect |
| train with multiple partitionings instead of single partitioning (ablation) | base (L, m) · Im2GPS3k test | accuracy within 25 km: 24.9 -> 25.5 | Improved |
| replace VGG16 with ResNet-101 backbone (ablation) | base (L, m) · Im2GPS3k test | accuracy within 25 km: 16.5 -> 24.9 | Improved |
| use mean GPS of training images in predicted cell instead of geographical cell center (ablation) | base-vgg (L, m) · Im2GPS3k test | accuracy within 25 km: 14.6 -> 16.5 | Improved |
| restrict a single-partitioning network to scene concepts (ISNs) (ablation) | base (L, m) · Im2GPS3k test | accuracy | No clear effect |
| add scene classification as auxiliary task (Multi-Task Network) (ablation) | base (M, f*) · Im2GPS3k test | accuracy | No clear effect |
Closest papers by meaning
- GeoRanker: Distance-Aware Ranking for Worldwide Image Geolocalization (NeurIPS 2025 · similarity 0.88)
- G3: An Effective and Adaptive Framework for Worldwide Geolocalization Using Large Multi-Modality Models (NeurIPS 2024 · similarity 0.87)
- Around the World in 80 Timesteps: A Generative Approach to Global Visual Geolocation (CVPR 2025 · similarity 0.87)
- GeoRouter: Dynamic Paradigm Routing for Worldwide Image Geolocalization (NeurIPS 2026 · similarity 0.87)
- Vision-Language Reasoning for Geolocalization: A Reinforcement Learning Approach (AAAI 2026 · similarity 0.86)
- PIGEON: Predicting Image Geolocations (CVPR 2024 · similarity 0.86)
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.
