SAM2Matting：通用圖像與視頻摳圖

摘要

儘管影像摳像（image matting）已取得顯著進展，但影片摳像（video matting）仍具挑戰性，主因在於高階追蹤（需理解全幀語境）與低階摳像（專注於極細微細節）之間存在天生鴻溝。現有方法依賴昂貴且範疇狹隘的影片摳像數據集，可能限制域外泛化能力並削弱追蹤穩健性。我們重新思考此範式，提出SAM2Matting——一個從追蹤器到摳像的框架，將視覺物件分割追蹤器（VOS tracker）升級為高保真影片摳像系統。具體而言，該框架透過引入區域提案橋接模組（region-proposal bridge）與專用摳像頭（matting heads）來解耦任務，使未受妥協的追蹤器處理時序一致性，同時由摳像元件解析細粒度細節。值得注意的是，SAM2Matting僅以影像進行訓練，卻在影片摳像中創下最新最佳表現，支援多樣提示類型，維持強大時序一致性，並在人物為中心與開放場景中展現穩健的泛化能力。

English

Despite impressive advances in image matting, video matting remains challenging due to the inherent gap between high-level tracking, which requires frame-wise understanding, and low-level matting, which focuses on extremely fine-grained details. Existing methods attempt this with expensive and narrowly-scoped video matting datasets, which may limit out-of-domain generalization and compromise tracking robustness. We rethink the paradigm with SAM2Matting, a tracker-to-matting framework that advances VOS trackers to high-fidelity video matting. Specifically, it decouples the task by enhancing a foundational tracker (e.g., SAM2, SAM3) with a region-proposal bridge and dedicated matting heads, enabling the uncompromised tracker to handle temporal consistency while the matting components resolve fine-grained details. Notably, despite being trained only on images, SAM2Matting establishes new state-of-the-art performance on video matting, supports diverse prompt types, maintains strong temporal consistency, and demonstrates robust generalization across both human-centric and in-the-wild scenarios.