PROJECT TITLE :

Online Kernel Slow Feature Analysis for Temporal Video Segmentation and Tracking - 2015

ABSTRACT:

Slow feature analysis (SFA) is a Dimensionality Reduction technique that has been linked to how visual brain cells work. In recent years, the SFA was adopted for pc vision tasks. In this paper, we tend to propose an actual kernel SFA (KSFA) framework for positive definite and indefinite kernels in Krein house. We have a tendency to then formulate an online KSFA that employs a reduced set expansion. Finally, by utilizing a special quite kernel family, we formulate actual online KSFA for that no reduced set is needed. We have a tendency to apply the proposed system to develop a SFA-based modification detection algorithm for stream data. This framework is employed for temporal video segmentation and tracking. We tend to test our setup on artificial and real information streams. When combined with an Online Learning tracking system, the proposed modification detection approach improves upon tracking setups that do not utilize amendment detection.