StreamGVE: Training-Free Video Editing via Few-Step Streaming Video Generation

TL;DR

StreamGVE introduces a novel noise-to-data video editing approach using pre-trained streaming models, enabling high-quality editing with minimal steps and reduced computational cost.

Contribution

It proposes a new noise-to-data paradigm for video editing, with dual-branch fast sampling, source-oriented guidance, and visual prompting, improving efficiency and flexibility.

Findings

Outperforms existing methods across various video editing tasks.

Achieves high-quality results with few-step sampling and minimal time.

Demonstrates robustness and generalizability across models.

Abstract

Although existing video editing methods are generally feasible, they often require many costly iterations and still struggle to deliver high-quality yet satisfying editing results. We attribute this limitation to the prevalent data-to-data paradigm, which is less compatible with modern generative models than noise-to-data generation. To address this gap, we revisit video editing from a noise-to-data perspective and propose Streaming-Generation-based Video Editing (StreamGVE), which preserves few-step sampling while seamlessly injecting source-video conditions. Built on pre-trained streaming generation models, StreamGVE introduces dual-branch fast sampling with a self-attention bridge and cross-attention grounding/boosting to satisfy both sampling and conditioning requirements. We further propose source-oriented guidance to improve target-generation quality, and a visual prompting strategy to enhance editing flexibility and practicality. The method is effective, robust, and generalizable across different models. Extensive experiments on diverse video editing tasks show that StreamGVE consistently outperforms existing approaches, even in few-step settings with minimal time cost.