Work

Trajectory Clustering Using Affinity Propagation with Trajectory Entropy Descriptor

Year: 2016

Type: conference-paper

Abstract: To cluster similar trajectories in surveillance videos, weconsider using trajectory entropy descriptors, which arecomputed from trajectories of foreground objects. Based onthe descriptors, we can assess the distances betweentrajectories with different lengths and transfer the distancesto a similarity matrix for affinity propagation clustering. Inthe experiments, we have shown the effectiveness andefficiency of the proposed method for trajectory clusteringcompared to the state-of-the-art methods.

Source: IIAE International Conference on Industrial Application Engineering (ICIAE)

DOI: 10.12792/iciae2016.096

Authors: Wei-Yun Huang, Wei-Cheng Wang, Chih-Li Chang, Wei-An Wang, Chun-Rong Huang

Institutions: National Chung Hsing University, National Cheng Kung University

Language: English

Topic: Video Surveillance and Tracking Methods

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: Ministry of Science and Technology, Taiwan

Awards: G2494647904, Spatial-Temporal Trajectory Descriptor Based Unsupervised Trajectory Event Analysis and Visualization