Dynamic Video Face Transformation Using Multilinear and Autoregressive Models

Research output: Contribution to conferencePaperpeer-review


In this paper we present a prototype system for altering perceived attributes of faces in video sequences, such as the apparent age, sex or emotional state. The system uses multilinear models to decompose the parameters coding for each frame into separate pose and identity parameters. The multilinear model is learnt automatically from the training video data. Statistical models of group identity are then used to alter the identity parameters from one group to another (e.g. from male to female). An autoregressive model is learnt from the pose parameters, and this is applied to alter the dynamics. We have tested our system on a small dataset (for altering apparent gender) with encouraging preliminary results.
Original languageEnglish
Number of pages4
Publication statusPublished - 06 Sept 2011
EventTheory and Practice of Computer Graphics - Warwick University, UK
Duration: 06 Sept 201108 Sept 2011


ConferenceTheory and Practice of Computer Graphics
CityWarwick University, UK
Period06 Sept 201108 Sept 2011


Dive into the research topics of 'Dynamic Video Face Transformation Using Multilinear and Autoregressive Models'. Together they form a unique fingerprint.

Cite this