音频驱动全身视频生成模型 夸克与浙江大学联合开源OmniAvatar

Core Insights - The article highlights the launch of OmniAvatar, an innovative audio-driven full-body video generation model developed by Quark Technology Team in collaboration with Zhejiang University [1] Group 1: Product Features - OmniAvatar requires only a single image and an audio clip to generate corresponding videos, significantly enhancing lip-sync detail and the fluidity of full-body movements [1] - The model allows for precise control over character poses, emotions, and scenes through the use of prompt words [1]