2D Model-Based Tracking of Humans in Single View Sequences

This work contributes to detection and tracking of walking or running humans in surveillance video sequences. We propose a 2D model-based approach to the whole body tracking in a video sequence captured from a single camera view. An extended six-link biped human model is employed. We assume that a static camera observes the scene horizontally or obliquely. Persons can be seen from a continuum of views ranging from a lateral to a frontal one. We do not expect humans to be the only moving objects in the scene and to appear at the same scale at different image locations.

• Person Walking With Backpack, Charles Square [AVI], 4.7MB
• Person Walking With Plant, Wallenstein Garden [AVI], 1.8MB
• Walking Person, orig. video by A. Fexa, Part 1 [AVI], 1.8MB
• Walking Person,  orig. video by A. Fexa, Part 2 [AVI], 1.3MB
• You may run your algorithms on the original sequences for comparison: [AVI 1], [AVI 2], [AVI 3], [AVI 4]

References:

1. Korč, F. and Hlaváč, V.: Detection and Tracking of Humans in Single View Sequences Using 2D Articulated Model
   Book chapter in Human Motion - Understanding, Modeling, Capture and Animation
   Springer's series: Computational Imaging and Vision, Vol. 36, 2007
   [PDF], [BibTeX]
2. Korč, F: 2D Model-Based Tracking of Humans in a Single View Sequence
   Master Thesis, advisor V. Hlaváč, Czech Technical University in Prague, Czech Republic, January 2006.