InfinityEdit：利用輕量級編輯點火適配器實現無限影片編輯

摘要

藉助大型預訓練模型，現有方法已有效改善基於指令的影片編輯。然而，其中大多數依賴於就地編輯假設：它們在固定的時間跨度內，將編輯後的影片與給定的來源片段逐幀對齊。這種模式無法適用於開放式串流，例如對實況遊戲進行重新風格化，或對進行中的鏡頭套用運鏡操作。在這些情況下，編輯必須擴展至隨到達的新增幀，而非套用於靜態輸入片段。本文研究此設定，並將其命名為無限影片編輯：給定前置片段與編輯請求，模型必須生成下一個片段，在套用所請求編輯的同時延續該串流。此過程會隨著編輯指令的無限序列到來而不斷重複。此任務帶來兩項挑戰：編輯必須是忠實的延續，而非逐幀重寫；且生成品質必須在編輯累積時保持穩定。為解決這些問題，我們首先設計了一套用於無限影片編輯的資料收集流程。基於所收集的資料，我們提出 InfinityEdit，這是一個輕量級編輯適配器，賦予串流影片生成器無限的編輯能力。該適配器包含三個注意力模組：歷史交叉注意力使用輸入幀引導去噪幀；時間因果自注意力僅讓時間線索從較早幀流向較晚幀；編輯交叉注意力則將編輯請求注入生成過程。在推論期間，適配器僅在編輯請求到達的區塊中啟動，後續區塊則由原始模型以重設錨定幀的方式生成。此方案在套用編輯的同時，保留了原始模型的無限生成能力。大量實驗顯示，InfinityEdit 能在每次編輯下忠實地延續串流，並在無限的編輯序列中保持穩定。

English

With large pretrained models, existing methods have effectively improved instruction-based video editing. However, most of them rely on an in-place editing assumption. They align the edited video with the given source clip frame by frame over a fixed time span. This pattern fails for open-ended streams, e.g., restyling a live game or applying a camera move to an ongoing shot. In such cases, edits must extend to future frames as they arrive, rather than be applied to a static input clip. In this paper, we study this setting and name it infinite video editing: given a preceding segment and an edit request, a model must generate the next segment that continues the stream while applying the requested edit. This process repeats as an unbounded sequence of edit instructions arrives. This task brings two challenges: the edit must be a faithful continuation rather than a frame-wise rewrite, and generation quality must remain stable as edits accumulate. To address them, we first design a data-collection pipeline for infinite video editing. Based on the collected data, we propose InfinityEdit, a lightweight edit adapter that equips a streaming video generator with unbounded editing ability. The adapter contains three attention modules. History cross-attention guides the denoising frames using the input frames. Temporal causal self-attention keeps temporal cues flowing only from earlier frames to later ones. Edit cross-attention injects the edit request into generation. During inference, the adapter is activated only in the chunk where an edit request arrives. Subsequent chunks are generated by the original model with a reset anchor frame. This scheme applies the edit while preserving the original model's infinite generation ability. Extensive experiments show that InfinityEdit faithfully continues the stream under each edit, and stays stable over unbounded edit sequences.