Disclosed is a processing method used in a video conference system having a local terminal device and at least one remote terminal device, comprising a feature detection step of carrying out a feature detection in the local terminal device and the remote terminal device, respectively; a conference mode setting step of setting a conference mode for the local terminal device and the remote terminal device, respectively; and a video encoding step of dividing a video frame in the local terminal device and the remote terminal device, and then pre-processing the divided areas so as to encode the pre-processed areas, respectively.