Imperatives for Virtual Humans

TL;DR

This paper introduces a framework that uses structured English to author more natural and intelligent virtual human behaviors, focusing on object properties, quantifiers, and spatial relations to improve realism.

Contribution

It presents a novel natural language-inspired framework for authoring virtual human behaviors, enhancing flexibility and realism in virtual character simulation.

Findings

Framework enables natural language-based behavior authoring

Improves realism and flexibility of virtual human behaviors

Supports structured English input for behavior specification

Abstract

Seemingly since the inception of virtual humans, there has been an effort to make their behaviors more natural and human-like. In additions to improving movement's visual quality, there has been considerable research focused on creating more intelligent virtual characters. This paper presents a framework inspired by natural language constructs that aims to author more reasonable virtual human behaviors using structured English input. We focus mainly on object types and properties, quantifiers, determiners, and spatial relations. The framework provides a natural, flexible authoring system for simulating human behaviors.