Crafting Realistic Virtual Humans: Unveiling Perspectives on Human Perception, Crowds, and Embodied Conversational Agents

Rubens Montanha,Victor Araujo,Paulo Knob,Greice Pinho,Gabriel Fonseca,Vitor Peres,Soraia Raupp Musse

DOI: https://doi.org/10.1109/SIBGRAPI59091.2023.10347175

2023-12-06

Abstract:Virtual Humans (VHs) were first developed more than 50 years ago and have undergone significant advancements since then. In the past, creating and animating VHs was a complex task. However, contemporary commercial and freely available technology now empowers users, programmers, and designers to create and animate VHs with relative ease. These technologies have even reached a point where they can replicate the authentic characteristics and behaviors of real actors, resulting in VHs that are visually convincing and behaviorally lifelike. This paper explores three closely related research areas in the context of virtual humans and discusses the far-reaching implications of highly realistic characters within these domains.

Human-Computer Interaction,Graphics

What problem does this paper attempt to address?

### What problems does this paper attempt to solve?

This paper aims to explore and solve the key problems of virtual humans (Virtual Humans, VHs) in three closely - related research fields: **human perception, crowd simulation and embodied conversational agents (Embodied Conversational Agents, ECAs)**. Specifically, the paper attempts to solve the following problems:

1. **Improve the fidelity of virtual humans**:
- The paper discusses how to create more realistic virtual humans through the latest technological means (such as high - fidelity modeling, animation techniques and machine learning, etc.). This includes visual realism and behavioral naturalness.
- Formula representation: Suppose \( R \) represents the fidelity of virtual humans and \( T \) represents the technological level, then it can be simply represented as:
\[
R = f(T)
\]
where \( f \) is an increasing function, meaning that the higher the technological level, the higher the fidelity of virtual humans.

2. **Understand and optimize human perception of virtual humans**:
- Research the "uncanny valley" effect, that is, when virtual humans are too close to the real but not completely similar, it may cause discomfort to people. The paper explores how to reduce this discomfort by improving the design of virtual humans.
- Analyze the influence of different factors (such as facial expressions, movement fluency, etc.) on human perception and propose a computational model to evaluate the comfort of virtual humans.
- Formula representation: Suppose \( C \) represents the comfort of virtual humans, \( F \) represents the realism of facial expressions, and \( M \) represents the movement fluency, then it can be represented as:
\[
C = g(F, M)
\]
where \( g \) is a comprehensive evaluation function that takes multiple perceptual factors into account.

3. **Enhance the performance of virtual humans in practical applications**:
- In terms of **crowd simulation**, the paper explores how to create and simulate large - scale crowd behaviors for application in urban planning, emergency evacuation and other fields.
- In terms of **embodied conversational agents**, the paper studies how to enable virtual humans to have more natural language and non - language communication abilities for application in health consultation, skill training and other scenarios.
- Formula representation: Suppose \( P \) represents the accuracy of crowd simulation and \( A \) represents the complexity of individual behaviors, then it can be represented as:
\[
P = h(A)
\]
where \( h \) is a function that describes the influence of the complexity of individual behaviors on the overall simulation accuracy.

### Summary

In general, this paper attempts to solve the challenges of virtual humans in perception, behavior and application through interdisciplinary research methods and technological means, thereby promoting the development of virtual human technology and making it widely used in more fields.