Avatar V：擴展視頻參考的虛擬形象視頻生成

摘要

生成不仅在外观上相似、更在行为上可识别的虚拟形象视频——即忠实复现目标人物的说话节奏、手势倾向与表情动态——仍是一项开放挑战。现有方法主要依赖单张静态图像作为条件，但此类图像提供的人物身份信息不足，且无法捕捉动态运动特征；此外，标准的像素级损失函数难以充分服务于决定虚拟形象逼真度的感知关键面部区域。我们提出Avatar V，一个面向生产规模的框架，通过视频参考条件化的身份建模来解决上述局限。该模型并非将身份信息压缩为固定大小的嵌入向量，而是直接以参考视频的完整标记序列为条件，通过注意力机制基于参考上下文学习复现静态身份属性（面部几何、皮肤纹理）与动态行为模式（说话节奏、微表情）。我们引入了稀疏参考注意力——一种非对称机制，能够以线性复杂度对任意长度的参考视频进行条件化；此外，还构建了运动表征流以实现闭环说话风格迁移，以及继承完整参考条件化的身份感知超分辨率精修器。上述技术依托于一个数据引擎，该引擎从5000万原始视频中精选了超过1亿条训练片段，并采用包含流匹配预训练、个性微调、两阶段蒸馏（10倍以上加速）及RLHF对齐的五阶段训练流程，部署在数千张GPU上。Avatar V可生成长度不限的1080p视频，在跨场景基准测试中实现了身份保持、唇形同步及生成质量方面最优的性能，在自动化指标与人工评估上均持续优于包括Seedance 2.0、Kling O3 Pro、Veo 3.1及OmniHuman 1.5在内的领先系统。

English

Generating avatar videos that are not merely visually similar to a target individual but behaviorally recognizable, faithfully reproducing their talking rhythm, gestural tendencies, and expression dynamics, remains an open challenge. Existing methods predominantly condition on single static images, which provide insufficient identity information and cannot capture dynamic motion traits, while standard pixel-level objectives underserve the perceptually critical facial regions that determine avatar fidelity. We present Avatar V, a production-scale framework that addresses these limitations through video-reference-conditioned identity modeling. Rather than compressing identity into fixed-size embeddings, the model conditions directly on the full token sequence of a reference video, learning to reproduce both static identity attributes (facial geometry, skin texture) and dynamic behavioral patterns (talking rhythm, micro-expressions) through attention over the reference context. We introduce Sparse Reference Attention, an asymmetric mechanism achieving linear-complexity conditioning on arbitrarily long references; a motion representation stream enabling closed-loop talking style transfer; and an identity-aware super-resolution refiner inheriting the full reference conditioning. These are supported by a data engine curating 100M+ training clips from 50M raw videos, and a five-stage training pipeline with flow matching pre-training, personality fine-tuning, two-phase distillation (>10x acceleration), and RLHF alignment, deployed across thousands of GPUs. Avatar V generates 1080p videos of unlimited duration, achieving state-of-the-art identity preservation, lip synchronization, and generation quality on our cross-scene benchmark, consistently outperforming leading systems including Seedance 2.0, Kling O3 Pro, Veo 3.1, and OmniHuman 1.5 in both automated metrics and human evaluation.