Output details
11 - Computer Science and Informatics
University of Birmingham
An adaptive coupled-layer visual model for robust visual tracking
<23>A novel approach for tracking objects, which undergo rapid and significant appearance changes. A new coupled-layer visual model that combines the target’s global and local appearance by probabilistically interlacing the two layers achieves a highly robust tracking. The tracker experimentally outperformed (on average) eleven state-of-the-art trackers on challenging sequences having smaller failure rate and better accuracy. Published at ICCV (CiteSeer IF rankings-in top 5% of all CS journals and conferences). The code has been made available to CV community. The work attracted industrial funding from a Californian startup company (Motrr, LLC). Extended version appeared in the top journal IEEE PAMI.