Digital humans are a highly realistic form of conversational computer agent. Eye gaze is a salient social cue that digital humans could use to facilitate rapport-building during conversations. However, eye gaze tendencies vary by gender and incorrect gaze patterns can have negative social implications. Analysis of observational data during human conversations can help inform the development of eye gaze models for digital humans. This study aimed to identify the eye gaze patterns of women dyads during a rapport-building conversation, and to evaluate the effect of different gaze patterns on rapport, trust, and psychological outcomes. 36 adult women (18 dyads) completed the Relationship Closeness Induction Task while wearing eye tracking glasses. Subjective rapport, trust, and psychological measures were collected. Gaze patterns of women were found to change as the conversation content became more intimate; specifically, gaze aversions for thinking (p=.042), turn-taking (p=.025), and intimacy modulation increased in duration (p=.012). Furthermore, gaze patterns were associated with perceptions of the conversation partner. Displaying fewer cognitive gaze aversions was associated with greater closeness (p=.029) and trust perceptions (p=.035). Longer periods of direct gaze while speaking was associated with greater rapport (p=.040). Results will inform the development of a humanlike gaze model for female digital humans during intimate conversations and may be applicable to social robots.

Our behaviour emerges as the result of many systems interacting at different scales, from low level biology to high level social interaction. Is it possible to create naturalistic explanatory models which can integrate these factors? We describe the general approach and design of a framework to create autonomous expressive embodied models of behaviour based on affective and cognitive neuroscience theories, sharing similar goals with current research in developmental robotics and embodied cognition, but with emphasis on computer graphics and face to face interaction. We aim to create a “large functioning sketch” of several fundamental aspects of behaviour affecting the face to explore emergence from interaction of low level and high level systems in a top-down bottom up approach.

We present a method to acquire the reflectance field of a human face and use these measurements to render the face under arbitrary changes in lighting and viewpoint. We first acquire images of the face from a small set of viewpoints under a dense sampling of incident illumination directions using a light stage. We then construct a reflectance function image for each observed image pixel from its values over the space of illumination directions. From the reflectance functions, we can directly generate images of the face from the original viewpoints in any form of sampled or computed illumination. To change the viewpoint, we use a model of skin reflectance to estimate the appearance of the reflectance functions for novel viewpoints. We demonstrate the technique with synthetic renderings of a person's face under novel illumination and viewpoints.

An anatomically detailed 3-D computer graphic model of the eye and surrounding face within a virtual environment has been implemented for use in a surgical simulator. The simulator forms part of a teleoperated micro-surgical robotic system being developed for eye surgery. The model has been designed to both visually and mechanically simulate features of the human eye by coupling computer graphic realism with finite element analysis.