Evolving networks for group object motion estimation

• Published in: IET Seminar on Target Tracking and Data Fusion: Algorithms and Applications pp. 97 - 106
• Main Authors: Gning, A, Mihaylova, L, Maskell, S, Sze Kim Pang, Godsill, S
• Format: Conference Proceeding
• Language: English
• Published: Stevenage IET 2008
• Subjects: Combinatorial mathematics; Computer vision and image processing techniques; Monte Carlo methods; Optical, image and video signal processing; evolving graph network model; Monte Carlo methods; sequential Monte Carlo framework; graph theory; motion estimation; object detection; group object motion estimation; tracking; group object tracking technique; automatic graph structure initialisation
• ISBN: 0863419100; 9780863419102
• DOI: 10.1049/ic:20080061

This paper proposes a technique for group object motion estimation based on evolving graph networks. The main novelty over alternative group tracking techniques stems from learning the network structure for the group. An algorithm is proposed for automatic graph structure initialisation, incorporation of new nodes and unexisting nodes removal in parallel with the edge update. This evolving graph model is combined with the sequential Monte Carlo framework and its effectiveness is illustrated over a complex scenario for group motion estimation in urban environment. Results with merging, splitting and crossing of the groups are presented with high estimation accuracy.