Revisiting CNNs for Trajectory Similarity Learning
Summary: Revisits CNNs for trajectories, arguing local similarity matters more than long-range dependency and proposing ConvTraj with 1D convs for sequential patterns and 2D convs for geo-distribution. With theoretical support, ConvTraj achieves SOTA accuracy and large speedups (240× training, 2.16× inference on 1.6M Porto trajectories). (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Zhihao Chang (Zhejiang University)
- 2. Linzhu Yu (Zhejiang University)
- 3. Huan Li (Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security; Zhejiang University)
- 4. Sai Wu (Zhejiang University)
- 5. Gang Chen (Zhejiang University)
- 6. Dongxiang Zhang (Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security; Zhejiang University)
BibTeX Citation
@article{chang_vldb25,
title = {{Revisiting CNNs for Trajectory Similarity Learning}},
author = {Chang, Zhihao and Yu, Linzhu and Li, Huan and Wu, Sai and Chen, Gang and Zhang, Dongxiang},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {4},
pages = {1013--1021},
doi = {10.14778/3717755.3717762},
url = {https://doi.org/10.14778/3717755.3717762},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 6 of 6 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 221 | Robust and Fast Similarity Search for Moving Object Trajectories | 2005 | SIGMOD | 0.00024224879 |
| 303 | On The Marriage of Lp-norms and Edit Distance | 2004 | VLDB | 0.00021956234 |
| 2,046 | Distributed Trajectory Similarity Search | 2017 | VLDB | 9.260657e-05 |
| 2,416 | DITA: Distributed In-Memory Trajectory Analytics | 2018 | SIGMOD | 8.6068712e-05 |
| 4,803 | Subtrajectory Clustering: Models and Algorithms | 2018 | PODS | 6.5016379e-05 |
| 6,953 | Trajectory Similarity Measurement: An Efficiency Perspective | 2024 | VLDB | 5.7303405e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 5,071 | Fast Subtrajectory Similarity Search in Road Networks under Weighted Edit Distance Constraints | 2020 | VLDB |
| 2 | 6,579 | A Deep Generative Model for Trajectory Modeling and Utilization | 2023 | VLDB |
| 3 | 6,285 | Trajectory Similarity Join in Spatial Networks | 2017 | VLDB |
| 4 | 11,456 | Efficient Non-Learning Similar Subtrajectory Search | 2023 | VLDB |
| 5 | 10,849 | SimRN: Trajectory Similarity Learning in Road Networks based on Distributed Deep Reinforcement Learning | 2025 | VLDB |
| 6 | 4,316 | Searching Trajectories by Locations - An Efficiency Study | 2010 | SIGMOD |
| 7 | 10,953 | SIMformer: Single-Layer Vanilla Transformer Can Learn Free-Space Trajectory Similarity | 2025 | VLDB |
| 8 | 1,885 | Discovery of Convoys in Trajectory Databases | 2008 | VLDB |
| 9 | 5,997 | Efficient and Effective Similar Subtrajectory Search with Deep Reinforcement Learning | 2020 | VLDB |
| 10 | 6,953 | Trajectory Similarity Measurement: An Efficiency Perspective | 2024 | VLDB |