A virtual human generation method combining user emotional preferences with implicit reconstruction

Key Points

• The aim is to develop a method for generating virtual humans that align with user emotional preferences.
• Developed a virtual human generation algorithm incorporating user emotional inputs.
• Conducted tests to evaluate emotional responses from users interacting with generated characters.
• Employed implicit reconstruction techniques for realistic depiction of emotional states.
• Participants reported a higher emotional connection with virtual humans tailored to their preferences (p<0.05).
• Implicit reconstruction enhanced the realism of emotional expressions, leading to more favorable user feedback.
• Overall satisfaction with virtual human interactions increased significantly compared to traditional methods.

Abstract