opencv - Visual tracking of a known object/person -
say identify , bounding-box-track person within given video. if there advantage have prior knowledge of person looks -- e.g., body features, clothes, etc. (i.e., extensive offline training possible), seems should able exploit make tracking more accurate , drift-free.
in literature, there name class of cv algorithms/research concerned different problem (as opposed standard tracking of unknown target)?
Comments
Post a Comment