# Avatar V：扩展视频参考的虚拟人视频生成

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-06-11 08:00
- AIHOT 分数：50
- AIHOT 链接：https://aihot.virxact.com/items/cmqemsqua03pdslun27cykrxc
- 原文链接：https://arxiv.org/abs/2606.13872

## AI 摘要

Avatar V 是一个生产级框架，通过视频参考条件建模替代静态图像驱动。模型直接对参考视频的完整 token 序列进行注意力计算，同时重现静态身份和动态行为（如说话节奏、微表情）。核心技术包括线性复杂度的稀疏参考注意力、支持闭环风格迁移的运动表示流、继承全参考条件的身份感知超分精炼器。数据引擎从 5000 万原始视频中筛选出 1 亿以上训练片段，经 flow matching 预训练、个性微调、两阶段蒸馏（>10 倍加速）和 RLHF 对齐等五阶段训练，部署于数千 GPU。可生成无限时长 1080p 视频，在跨场景基准上保持最优的身份保留、唇同步和生成质量，全面超越 Seedance 2.0、Kling O3 Pro、Veo 3.1 和 OmniHuman 1.5。

## 正文

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.