Characteristic facial retargeting

Abstract

Facial motion retargeting has been developed mainly in the direction of representing high fidelity between a source and a target model. We present a novel facial motion retargeting method that properly regards the significant characteristics of target face model. We focus on stylistic facial shapes and timings that reveal the individuality of the target model well, after the retargeting process is finished. The method works with a range of expression pairs between the source and the target facial expressions and emotional sequence pairs of the source and the target facial motions. We first construct a prediction model to place semantically corresponding facial shapes. Our hybrid retargeting model, which combines the radial basis function (RBF) and kernel canonical correlation analysis (kCCA)-based regression methods copes well with new input source motions without visual artifacts. 1D Laplacian motion warping follows after the shape retargeting process, replacing stylistically important emotional sequences and thus, representing the characteristics of the target face.

Evaluation of a Facial Animation Authoring Pipeline Seamlessly Supporting Performance Capture and Manual Key-pose Editing

“…A quantitative evaluation is based on the computation of a numerical difference between the source and the target expressions. Previous results, whether qualitative [23] or quantitative [26], tell us that a perfect match between the source and the target is never achieved. The animation pipeline presented in this work is evaluated, qualitatively, by measuring the recognizability of the six basic facial emotions defined by Ekman [1].…”

“…The animation pipeline presented in this work is evaluated, qualitatively, by measuring the recognizability of the six basic facial emotions defined by Ekman [1]. In a previous work, Song et al [26] used facial emotions like expressing surprise, crying, and laughing to train the retargeting routine. In our work, we exploit emotion transfer to evaluate the retargeting.…”

2D to 3D space transformation for facial animation based on marker data

“…The value of function ф depends on only upon the distance from its center and thus is called radial (6). This radial is the distance between marker points on the face of a 2D facial image.…”

“…Reference [11] explained that manual editing is always necessary: 1) the movement is expected from the 3D model is not necessarily the same/similar to the movement of people, 2) 3D face model has a proportion of/morphology that is different from the human face, 3) the placement of markers in motion capture is not always fixed and precisely at a points from day to day. Therefore, generating good facial animation is one of the most important factors in the creation of appealing, believable, and attractive characters [6].…”

“…Motion that occurs in 3D face models is scaled According to the relative scale between the source and the target [4] [6]. Marker points of the source face define the source space while the feature points of the targets face defining the space targets [7] [8].…”