Evaluasi Sintesis Ekspresi Wajah Realistik pada Sistem Animasi Wajah 3D dengan Teknologi Motion Capture

Abstract

The human face has a unique shape and size, as well as a 3D character face model. The facial animation of 3D virtual characters is mostly done manually by moving the rigging in each frame. The more characters used, the more production costs that must be incurred. The absence of a cheap facial motion transfers system is also one of the reasons why not many studios are using motion capture technology in Indonesia.

This research will evaluate the implementation of a facial expression synthetically using motion capture technology built from radial basis function (RBF) as a method of marker transfer as a reference for rigging movement in point cluster system. Testing is done by approaching facial expressions according to FACS theory and questionnaire of synthesis results.

The experimental results show that according to FACS theory the requirement of expression formation has been fulfilled by referring to changes in facial features, but the implementation is not always able to describe perfectly the desired condition, namely the average percentage of faces easily recognizable by 35.53%. Therefore, the influence of animators in the control of micro expression improvements or the addition of exaggeration principle elements in the manufacture of facial animation is very important to produce facial expressions that are easily recognized by the audience.