AnimateDiff-Lightning：跨模型擴散蒸餾

AnimateDiff-Lightning: Cross-Model Diffusion Distillation

March 19, 2024

作者: Shanchuan Lin, Xiao Yang

cs.AI

摘要

我們提出了 AnimateDiff-Lightning 用於快速生成影片。我們的模型採用漸進式對抗擴散蒸餾，實現了在少步驟影片生成方面的新最先進技術。我們討論了我們對其進行的修改，以適應影片模式。此外，我們提議同時蒸餾多個基礎擴散模型的概率流，從而產生一個具有更廣泛風格兼容性的單一蒸餾運動模組。我們很高興釋出我們蒸餾的 AnimateDiff-Lightning 模型供社群使用。

English

We present AnimateDiff-Lightning for lightning-fast video generation. Our model uses progressive adversarial diffusion distillation to achieve new state-of-the-art in few-step video generation. We discuss our modifications to adapt it for the video modality. Furthermore, we propose to simultaneously distill the probability flow of multiple base diffusion models, resulting in a single distilled motion module with broader style compatibility. We are pleased to release our distilled AnimateDiff-Lightning model for the community's use.