Evaluating the perceptual realism of animated facial expressions

Abstract

The human face is capable of producing an astonishing variety of expressions-expressions for which sometimes the smallest difference changes the perceived meaning considerably. Producing realistic-looking facial animations that are able to transmit this degree of complexity continues to be a challenging research topic in computer graphics. One important question that remains to be answered is: When are facial animations good enough? Here we present an integrated framework in which psychophysical experiments are used in a first step to systematically evaluate the perceptual quality of several different computer-generated animations with respect to real-world video sequences. The first experiment provides an evaluation of several animation techniques, exposing specific animation parameters that are important to achieve perceptual fidelity. In a second experiment, we then use these benchmarked animation techniques in the context of perceptual research in order to systematically investigate the spatiotemporal characteristics of expressions. A third and final experiment uses the quality measures that were developed in the first two experiments to examine the perceptual impact of changing facial features to improve the animation techniques. Using such an integrated approach, we are able to provide important insights into facial expressions for both the perceptual and computer graphics community.

Supramodal Representation of Emotions

“…The effectiveness of virtual characters for emotion recognition tasks has been successfully validated in patient and control populations (Dyck et al, 2008(Dyck et al, , 2010Wallraven et al, 2008). We used dynamic virtual characters (avatars) exhibiting angry, neutral, and happy facial emotions and combined them with pseudowords with angry, neutral, and happy prosody in congruent and incongruent trials, assuring standardized facial expressions and perfect lip-speech synchronization.…”

How real is reality? A perceptually motivated system for quantifying visual realism in digital images

“…Many of these techniques for image classification were unsuitable for use in this project because they are a) too obscure and cannot be generalized to a large range of images [23][24][25][26] or b) based on statistical modeling of characteristics beyond human perception [17,19,[27][28][29][30][31][32][33].…”

Investigating Facial Animation Production through Artistic Inquiry

“…At this stage, the practitioner considered contributing factors and experimented with expression manipulation using the tools of traditional animation practices. Contributing factors primarily consisted of existing research into facial expression generation and perception [7][12] [22][25] [33], the artistic depiction of facial expressions [9] [17], and traditional animation guides [32] [35]. In particular, the images of peak emotional expressions collected by Ekman and Friesen [7] were useful when the facial shapes were being constructed.…”

“…Additionally, spatiotemporal facial attributes have been manipulated and tested [33]. However, while most of these projects focus on the significance of spatial or temporal attributes to perception, none consider the artistic reasoning behind spatiotemporal facial animation choices.…”