Frame Context Packing and Drift Prevention in Next-Frame-Prediction Video Diffusion Models
This work addresses incremental improvements in video generation for AI and multimedia applications by enhancing context handling and stability in diffusion models.
The paper tackles the problem of error accumulation and limited context in next-frame-prediction video diffusion models by introducing FramePack, a method that compresses input frames with importance-based packing and drift prevention techniques, enabling inference with thousands of frames and reducing drift in video generation.
We present a neural network structure, FramePack, to train next-frame (or next-frame-section) prediction models for video generation. FramePack compresses input frame contexts with frame-wise importance so that more frames can be encoded within a fixed context length, with more important frames having longer contexts. The frame importance can be measured using time proximity, feature similarity, or hybrid metrics. The packing method allows for inference with thousands of frames and training with relatively large batch sizes. We also present drift prevention methods to address observation bias (error accumulation), including early-established endpoints, adjusted sampling orders, and discrete history representation. Ablation studies validate the effectiveness of the anti-drifting methods in both single-directional video streaming and bi-directional video generation. Finally, we show that existing video diffusion models can be finetuned with FramePack, and analyze the differences between different packing schedules.