Research map / Video / Video segmentation and tracking
Understanding Emerging Spatial Entities
Jinyoung Yeo, Jin-woo Park, Seung-won Hwang
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 full Ours instead of high-precision CLU | photo linking · Seattle | F1: .670 -> .742 | Improved |
| use full Ours instead of high-precision CLU | photo linking · Washington D.C. | F1: .546 -> .726 | Improved |
| use full Ours instead of high-precision CLU | photo linking · Taipei | recall: .414 -> .621 | Improved |
| use full Ours instead of high-precision CLU | photo linking · Seattle | precision: .929 -> .891 | Hurt |
| add multi-signal joint aggregation (C1) in place of independent tag recommendation and clustering (ablation) | Ours1 vs CLU+REC · Seattle and Washington D.C. | linking score separation | Improved |
| add synonym discovery using geo and image features (C2) in place of SYN (ablation) | Ours2 vs CLU+SYN · Seattle | SP@1: 0.91 -> 0.94 | Improved |
| use Ours2 instead of SYN for synonym extraction (ablation) | SE synonym extraction · Washington D.C. | SP@2: 0.50 -> 0.87 | Improved |
| use Ours2 instead of fuzzy matching (ablation) | SE synonym extraction · Seattle | SP@1: 0.33 -> 0.94 | Improved |
| combine geo-spatial and image features (lambda around 0.5) instead of either alone (ablation) | Ours2 · Seattle and Washington D.C. | linking performance | Improved |
| alternate C1 and C2 instead of iterating C2 alone (ablation) | Ours · Seattle and Washington D.C. | F1 over iterations | Improved |
| combine C1 and C2 (Ours converged) instead of C1 only (ablation) | Ours · Washington D.C. | F1: .627 -> .726 | Improved |
| use Ours instead of CLU for cross-city robustness | photo linking · Seattle vs Washington D.C. | F1 gap: ±.124 -> ±.016 | Improved |
Closest papers by meaning
- Unaligned Message-Passing and Contextualized-Pretraining for Robust Geo-Entity Resolution (AAAI 2025 · similarity 0.79)
- Spatial-Temporal Knowledge Distillation for Takeaway Recommendation (AAAI 2025 · similarity 0.78)
- Integrating Personalized Spatio-Temporal Clustering for Next POI Recommendation (AAAI 2025 · similarity 0.77)
- GeoMamba: Towards Multi-granular POI Recommendation with Geographical State Space Model (AAAI 2025 · similarity 0.77)
- Learn from View Correlation: An Anchor Enhancement Strategy for Multi-view Clustering (CVPR 2024 · similarity 0.76)
- GeoFidelity-Bench: Evaluating Block-Conditioned Geographic Fidelity of Street-View Generation (NeurIPS 2026 · similarity 0.76)
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.
