Online Moving Camera Background Subtraction


Recently several methods for background subtraction from moving camera were proposed. They use bottom up cues to segment video frames into foreground and background regions. Due to this lack of explicit models, they can easily fail to detect a foreground object when such cues are ambiguous in certain parts of the video. This becomes even more challenging when videos need to be processed online. We present a method which enables learning of pixel based models for foreground and background regions and, in addition, segments each frame in an online framework. The method uses long term trajectories along with a Bayesian ltering framework to estimate motion and appearance models. We compare our method to previous approaches and show results on challenging video sequences.

Publications

  • Ali Elqursh and Ahmed Elgammal. Online Motion Segmentation using Dynamic Label Propagation. ICCV 2013. Pdf
  • Ali Elqursh and Ahmed Elgammal. Online Moving Camera Background Subtraction. ECCV 2012. Pdf
  • Ali Elqursh and Ahmed Elgammal. Video Figure Ground Labeling. ICPR 2012. Pdf

Line-Based Relative Pose Estimation

We propose algorithms that solves the relative pose problem using lines.

Publications

  • Ali Elqursh and Ahmed Elgammal. Line-Based Relative Pose Estimation. CVPR 2011. Pdf, Code.
  • Ali Elqursh and Ahmed Elgammal. Single Axis Relative Rotation from Orthogonal Lines. ICPR 2012. Pdf