Extracting and Locating Temporal Motifs in Video Scenes
Using a Hierarchical Non Parametric Bayesian Model

Rémi Emonet, Jagannadan Varadarajan, Jean-Marc Odobez − CVPR 2011

This page is the entry point for the additional material associated with our cvpr paper. It contains details mentioned in the paper, more precisely:

• detailed equations for the inference process, cf section “5. Inference” of the paper.
  As many equations are involved, we provide a pdf file with equations.
• visuals for recovered motifs on synthetic documents, cf section “6.1. Validation on synthetic documents” of the paper.
  This is shown in a separate page for experiments on synthetic data.
• recovered motifs for more datasets and different parameters, cf section “6.2. Experiments on video data” and Fig. 6 of the paper.
  This is shown in a separate page experiments on video data.