DiffMove: Group Mobility Tendency Enhanced Trajectory Recovery via Diffusion Model PubID: pubid: 0000–0000 © 2025 IEEE
Qingyue Long, Can Rong, Huandong Wang, Shaw Rajib, Yong Li · 2025
Read DiffMove: Group Mobility Tendency Enhanced Trajectory Recovery via Diffusion Model PubID: pubid: 0000–0000 © 2025 IEEE by Qingyue Long, Can Rong, Huandong Wang, Shaw Rajib, Yong Li (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-18302.txt.
Description
Qingyue Long and co-authors present DiffMove, a method to fill gaps in movement trajectories. Their 2025 paper deals with location data that are sparse because devices record rarely or cover only part of a city. Sparse paths make a person's own history too thin to guide good reconstruction.
DiffMove borrows strength from the crowd: a graph built from all users' paths captures group movement trends. Those trends are blended into each location's features through graph embedding. The method also reads a user's habits from both past paths and the current one. A diffusion model then rebuilds the missing points. Tests on two real data sets beat prior methods.
Catalog record
| CREATOR | Qingyue Long, Can Rong, Huandong Wang, Shaw Rajib, Yong Li |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-18302