Avatar V: Scaling Video-Reference Avatar Video Generation
Quick Answer
Avatar V introduces a novel video-reference-conditioned identity modeling framework, achieving state-of-the-art avatar video generation with 1080p quality and superior identity preservation.
Quick Take
Avatar V introduces a novel video-reference-conditioned identity modeling framework, achieving state-of-the-art avatar video generation with 1080p quality and superior identity preservation. It outperforms leading systems like Seedance 2.0 and Kling O3 Pro in both automated metrics and human evaluations, utilizing a vast dataset of over 100 million training clips and advanced training techniques across thousands of GPUs.
Key Points
- Generates unlimited duration 1080p avatar videos with high fidelity.
- Utilizes Sparse Reference Attention for efficient long reference conditioning.
- Achieves superior lip synchronization and identity preservation metrics.
- Trained on a dataset of over 100 million clips from 50 million videos.
- Outperforms competitors in both automated metrics and human evaluations.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2606. 13872v1 Announce Type: new Abstract: 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.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval
A phase-aware LLM agent optimizes human-object interaction retrieval, outperforming Optuna TPE by 33.3% and VDTuner by 34.2% on the HICO-DET benchmark. This method enhances throughput by 15.3x over UniIR and demonstrates strong transferability across vector database management systems.
