Robust low rank dynamic mode decomposition for compressed domain crowd and traffic flow analysis | IEEE Conference Publication | IEEE Xplore

Robust low rank dynamic mode decomposition for compressed domain crowd and traffic flow analysis


Abstract:

In this paper, we develop a dynamic mode decomposition algorithm that is robust to both inlier and outlier noise in the data. One application of our algorithm is the iden...Show More

Abstract:

In this paper, we develop a dynamic mode decomposition algorithm that is robust to both inlier and outlier noise in the data. One application of our algorithm is the identification of multiple crowd or traffic flows from compressed video streams. Our method uses motion vectors that are readily available in the compressed bitstream, and do not require computationally expensive optical flow. These motion vectors are known to be very noisy, however, our algorithm is able to extract the underlying dynamical systems that define the flows. We formulate a rank regularized dynamic mode decomposition problem with total least squares constraints to estimate the Koopman modes of the motion dynamics. The estimated Koopman modes are then used to analyze the stability of the system and extract steady state and transient flows. We demonstrate the improved performance of our approach compared to state of the art schemes and illustrate it applicability in identifying transient and steady-state flows in real video sequences.
Date of Conference: 11-15 July 2016
Date Added to IEEE Xplore: 29 August 2016
ISBN Information:
Electronic ISSN: 1945-788X
Conference Location: Seattle, WA, USA

1. Introduction

The identification of motion flows within dense crowds of people in surveillance video sequences is an essential tool in crowd safety and crowd control tasks. Surveillance video of crowded scenes exhibit complex crowd behaviors even under normal situations. For example, crowd flows in large congested train stations can appear chaotic. However, it is often the case that low dimensional dynamical structures exist in the observed flow. Such structures are desirable to identify and segment from unstructured and transient flows. The automatic identification of different types of crowd flows aids in the monitoring and prediction of hazardous situations in crowded environments. Similarly, the identification of anomalous flows in traffic scenes helps management facilities predict and react to potential congestion.

References

References is not available for this document.