Geodesic active contours and level sets for the detection and tracking of moving objects

This paper presents a new variational framework for detecting and tracking multiple moving objects in image sequences. Motion detection is performed u sing a statistical framework for which the observed interframe difference d ensity function is approximated using a mixture model. This model is compos ed of two components, namely, the static (background) and the mobile (movin g objects) one. Both components are zero-mean and obey Laplacian or Gaussia n law. This statistical framework is used to provide the motion detection b oundaries. Additionally, the original frame is used to provide the moving o bject boundaries. Then, the detection and the tracking problem are addresse d in a common framework that employs a geodesic active contour objective fu nction. This function is minimized using a gradient descent method, where a flow deforms the initial curve towards the minimum of the objective functi on, under the influence of internal and external image dependent forces. Us ing the level set formulation scheme, complex curves can be detected and tr acked white topological changes for the evolving curves are naturally manag ed. To reduce the computational cost required by a direct implementation of the level set formulation scheme, a new approach named Hermes is proposed. Hermes exploits aspects from the well-known front propagation algorithms ( Narrow Band. Fast Marching) and compares favorably to them. Very promising experimental results are provided using real video sequences.