The merits are based on three factors: (1) The overall features of the trajectory clusters have more comprehensive and global information than single trajectory vectors; (2) the overall features used as the input of the classifier have better computational efficiency, independent of the system ...
The parameters αi and b are determined by a linearly constrained quadratic programming (QP) problem, which can be efficiently implemented through a sequence of smaller scale, sub-problem optimizations or an incremental scheme that adjusts the solution one training point at a time. Most of the ...