Graph-Based Correlated Topic Model for Trajectory Clustering in Crowded Videos
Abstract
This paper presents a graph-based correlated topic model (GCTM) to learn and analyse motion patterns by trajectory clustering in a highly cluttered and crowded environment. Unlike previous works that depend on scenes prior, we extract trajectories and apply a spatio-temporal graph (STG) to uncover the spatial and temporal coherence between the trajectories during the learning process. It advances the CTM by integrating a manifold-based clustering as initialization and iterative statistical inference as optimization. The output of GCTM are mid-level features that represent the motion patterns used later to generate trajectory clusters. Experiments on two different datasets show the effectiveness of the approach in trajectory clustering and crowd motion modelling.
Cite
Text
Al Ghamdi and Gotoh. "Graph-Based Correlated Topic Model for Trajectory Clustering in Crowded Videos." IEEE/CVF Winter Conference on Applications of Computer Vision, 2018. doi:10.1109/WACV.2018.00118Markdown
[Al Ghamdi and Gotoh. "Graph-Based Correlated Topic Model for Trajectory Clustering in Crowded Videos." IEEE/CVF Winter Conference on Applications of Computer Vision, 2018.](https://mlanthology.org/wacv/2018/ghamdi2018wacv-graph/) doi:10.1109/WACV.2018.00118BibTeX
@inproceedings{ghamdi2018wacv-graph,
title = {{Graph-Based Correlated Topic Model for Trajectory Clustering in Crowded Videos}},
author = {Al Ghamdi, Manal and Gotoh, Yoshihiko},
booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision},
year = {2018},
pages = {1029-1037},
doi = {10.1109/WACV.2018.00118},
url = {https://mlanthology.org/wacv/2018/ghamdi2018wacv-graph/}
}