Research map / Label-efficient and robust learning / Recognition and long-tail
MovSemCL: Movement-Semantics Contrastive Learning for Trajectory Similarity
Zhichen Lai, Hua Lu, Huan Li, Jialiang Li, Christian S. Jensen
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 movement-semantics encoding (ablation) | MovSemCL · Germany similarity search, DB 100K | mean rank: 4.122 -> 1.008 | Improved |
| add movement-semantics encoding (ablation) | MovSemCL · Porto similarity search, DB 100K | mean rank: 3.045 -> 1.005 | Improved |
| add curvature-guided augmentation (ablation) | MovSemCL · Germany similarity search, DB 100K | mean rank: 1.234 -> 1.008 | Improved |
| add curvature-guided augmentation (ablation) | MovSemCL · Porto similarity search, DB 100K | mean rank: 1.098 -> 1.005 | Improved |
| add hierarchical patch encoding (ablation) | MovSemCL · Germany similarity search, DB 100K | mean rank: 1.039 -> 1.008 | Improved |
| add hierarchical patch encoding (ablation) | MovSemCL · Porto similarity search, DB 20K | mean rank: 1.005 -> 1.002 | Improved |
| replace TrajCL with MovSemCL (patch-based attention) | TrajCL · Inference efficiency | latency (ms): 6.08 -> 3.44 | Improved |
| use MovSemCL instead of TrajCL for heuristic approximation | TrajCL · Porto heuristic approximation (HR@5) | HR@5: EDR 0.172 -> 0.207; EDwP 0.546 -> 0.536 | Mixed |
| use MovSemCL instead of TrajCL under heavy down-sampling | TrajCL · Porto down-sampling rate 0.5 | mean rank: 36.352 -> 9.951 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CLEAR — CLEAR: Cumulative LEARning for One-Shot One-Class Image Recognition (CVPR 2018) · better on 6 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- KGTS: Contrastive Trajectory Similarity Learning over Prompt Knowledge Graph Embedding (AAAI 2024 · similarity 0.88)
- TrajMamba: An Efficient and Semantic-rich Vehicle Trajectory Pre-training Model (NeurIPS 2025 · similarity 0.87)
- GTR: A General, Multi-View, and Dynamic Framework for Trajectory Representation Learning (ICML 2025 · similarity 0.86)
- Adapting to Observation Length of Trajectory Prediction via Contrastive Learning (CVPR 2025 · similarity 0.85)
- Holistic Semantic Representation for Navigational Trajectory Generation (AAAI 2025 · similarity 0.84)
- UniTraj: Learning a Universal Trajectory Foundation Model from Billion-Scale Worldwide Traces (NeurIPS 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.
