Trajectory-Based Video Retrieval Using Dirichlet Process Mixture Models

Abstract

In this paper, we present a trajectory-based video retrieval framework using Dirichlet process mixture models. The main contribution of this framework is four-fold. (1) We apply a Dirichlet process mixture model (DPMM) to unsupervised trajectory learning. DPMM is a countably infinite mixture model with its components growing by itself. (2) We employ a time-sensitive Dirichlet process mixture model (tDPMM) to learn trajectories' time-series characteristics. Furthermore, a novel likelihood estimation algorithm for tDPMM is proposed for the first time. (3) We develop a tDPMM-based probabilistic model matching scheme, which is empirically shown to be more error-tolerating and is able to deliver higher retrieval accuracy than the peer methods in the literature. (4) The framework has a nice scalability and adaptability in the sense that when new cluster data are presented, the framework automatically identifies the new cluster information without having to redo the training. Theoretic analysis and experimental evaluations against the state-of-the-art methods demonstrate the promise and effectiveness of the framework.

Outlier detection in traffic data based on the Dirichlet process mixture model

“…By this property, DPMM has recently become a widely discussed stochastic model in the applications of unsupervised clustering [14][15][16][17][18][19]. DPMM has been applied to many applications such as abnormal activity detection [14,18], scene categorization [16], tracking maneuvering targets [15], trajectory-based video retrieval [17], abnormal events [19] and motion segmentation [20]. In short, only one case [20]…”

Content Based Video Retrieval Systems

“…In order to speed up algorithm, authors only utilize the local motion embedded in the region-ofinterest as the query to retrieve data from MPEG bitstreams. In Trajectory-Based Video Retrieval Using Dirichlet Process Mixture Models [35], present a trajectory-based video retrieval framework using Dirichlet process mixture models. The main contribution of this framework is four-fold.…”

Interactive Image Segmentation Using Dirichlet Process Multiple-View Learning

“…In the single-view case, the base algorithm we utilize [31] is based on generative modeling using Dirichlet processes, which is shown to be powerful in many recent applications in computer vision [16]- [18], [21], [26]. The resulting framework in this paper is therefore referred to as Dirichlet process multiple-view learning (DPMVL), which provides systematic extensions to [31], such that the proposed model better exploits complementary image features by defining their distributions as mixture forms.…”