A framework for automatic and perceptually valid facial expression generation

Abstract

Facial expressions are facial movements reflecting the internal emotional states of a character or in response to social communications. Realistic facial animation should consider at least two factors: believable facial expression visual effect and valid facial movements. However, most research tends to separate these two issues. In this paper, we present a framework for generating facial animations considering both visual effect of facial expression and the dynamic factor. A facial expression mapping approach based on local geometry encoding is proposed, which encodes deformation in the 1-ring vector. This method is capable of mapping subtle facial movements without considering those shape and topological constraints. Facial expression mapping is achieved through three stepscorrespondence establishment, deviation transfer and movement mapping. Deviation is transferred to the conformal face space through minimizing the error function formed by the source neutral and the deformed face model related by those transformation matrices in 1-ring neighborhood. The transformation matrix in 1-ring neighborhood is independent of the face shape and mesh topology. Facial dynamics is then integrated with facial expressions for generating valid facial expressions. The dynamic factor was solved based on psychophysical methods. An application of the framework using various faces with different shapes and topological representations has been tested.

Realistic Facial Expression Reconstruction for VR HMD Users

“…To tackle these problems, a few recent studies resorted to the synthetic 3D facial data and yielded impressive results [34], [35]. This is in line with an interesting theory of parallel vision [36], [37] which discusses the significance of synthetic data [38], [39], [40], [41], [42] in addressing the problems of visual perception and understanding.…”

3D Face Tracking Using Stereo Cameras: A Review

“…This greatly hinders telemanipulation tasks, which differ with close object distances. There is a need to have natural vergence angles from human operator's eyes so that the right impressions of the depth of the remote environment are created [47].…”

“…However, stereo vision systems have tremendously improved this area of research by improving the quality of estimated geometry; hence the results that are observed have been relatively satisfactory, because texture information coding of the scenes has been limited [48]. 3D geometry has been estimated robustly by usage of time of flight and light coded cameras, which have served to estimate real time 3D geometry [47]. The downsides of these systems have been the limitations associated with low spatial resolutions.…”

Do fine feathers make a fine bird? The influence of attractiveness on fraud‐risk judgments by internal auditors

“…Stimuli from this database (for examples, see Figure ) have been and continue to be used to address a range of different research objectives. These include, for example, the development of frameworks for perceptually valid automatic facial expression generation (Yu, Garrod, Jack, & Schyns, ), the role of culture and motion with regard to perception of facial expressions (Jack, Garrod, Yu, Caldara, & Schyns, ; Richoz, Jack, Garrod, Schyns, & Caldara, ; Yu, Garrod, & Schyns, ), as well as the effect of race on face processing proficiency (Ramon et al, ).…”