Tensorial factorization methods for manipulation of face videos
S. Manikandan, R. Satheesh Kumar, C. V. Jawahar · 2006
This paper proposes the use of tensor factorization for manipulating videos of human faces. Decomposition of a video represented as a tensor into non-negative rank-1 factors results in sparse and separable factors equivalent to a local parts decomposition of the object in the video. Such decomposition can be used for tasks like expression transfer and face morphing. For instance, given a facial expression video it can be represented as a tensor which can then be factorized. The factors that best represent the expression can be identified which can then be transferred to another face video thus transferring the expression. Existing solutions to the problem of expression transfer require explicit modeling of the expression and its interaction with the underlying face content. Instead the method proposed here is purely appearance based and the results demonstrate that the proposed method is a simple alternative to the popular complex solution. A similar strategy has been used to morph a face image into a second face image. The resulting morph sequence is visually smooth indicating that the method can be used for generation of good quality morph sequence.