As S101, given an input video sequence X=x₁, x₂, . . . , such as data 121, a Face Detection & Facial Landmark Extraction module 122 first, at S102 determines one or a plurality of valid faces from each video frame x_i. In one embodiment, only the most prominent (e.g., largest) face are detected, and in another embodiment, all faces in the frame satisfying a condition (e.g., with a large enough size exceeding a threshold) are detected. At S103, for the j-th face in x_i, a set of facial landmarks are determined and a set of facial landmark features f_l,i,j, are computed correspondingly, which will be used by the decoder for restoring the j-th face in x_i. At S103, all facial landmark features of all faces are put together as F_l,i=f_l,i,1, f_l,i,2, . . . , data 124, which is encoded and transmitted by a Landmark Feature Compression & Transmission module 126. At S105, at the same time, for the j-th face in x_i, an Extended Face Area (EFA) can be computed by extending the bounding area (boundary as a rectangle, eclipse, or fine-grained segmentation boundary) of the original detected face to include additional hair, body parts, or even backgrounds. At S106 and S107, a set of EFA features f_b,i,j can be computed correspondingly, which will be used by the decoder for restoring the EFA of the j-th face in x_i. At S107, all EFA features of all faces are put together as F_b,i=f_b,i,1, f_b,i,2, . . . , data 125, which is encoded and transmitted by an EFA Compression & Transmission module 127.