Adaptive Feature Selection for Object Tracking with Particle Filter

Darshan Venkatrayappa (1) , Désiré Sidibé (2) , Fabrice Meriaudeau (2) , Philippe Montesinos (1)

1 LGI2P - Laboratoire de Génie Informatique et Ingénierie de Production
2 Le2i - Laboratoire Electronique, Informatique et Image [UMR6306]

Object tracking is an important topic in the field of computer vision. Commonly used color-based trackers are based on a fixed set of color features such as RGB or HSV and, as a result, fail to adapt to changing illumination conditions and background clutter. These drawbacks can be overcome to an extent by using an adaptive framework which selects for each frame of a sequence the features that best discriminate the object from the background. In this paper, we use such an adaptive feature selection method embedded into a particle filter mechanism and show that our tracking method is robust to lighting changes and background distractions. Different experiments also show that the proposed method outperform other approaches.

Mots clés: Tracking; Particle filter; Mean-shift filter; Feature selection

Domaines: Informatique [cs]; Vision par ordinateur et reconnaissance de formes [cs.CV]

Darshan Venkatrayappa, Désiré Sidibé, Fabrice Meriaudeau, Philippe Montesinos. Adaptive Feature Selection for Object Tracking with Particle Filter. International Conference on Image Analysis and Recognition, Oct 2014, Vilamoura, Algarve, Portugal. pp.395-402, ⟨10.1007/978-3-319-11755-3_44⟩. ⟨hal-01145635⟩