ChangeFlow——用於遙感變化檢測的潛在修正流

摘要

遥感变化檢測（RSCD）旨在定位同一地理區域兩張影像之間的變化。在實務中，變化遮罩常遵循區域級標註慣例，而非純粹基於局部外觀差異，因此具有上下文依賴性且偶有模糊性。多數先進方法採用逐像素判別式分類，每個輸入僅產出單一預測，無法將變化區域明確建模為一連貫整體。一個自然的替代方案是生成式模型，它能建模可行遮罩的分布，透過取樣來捕捉模糊性並促進全域一致性。然而，現有的生成式RSCD方法因像素空間生成的運算成本高昂及其條件化機制的複雜性，通常落後於強勢的判別式基線。為克服先前判別式與生成式方法的限制，我們提出ChangeFlow，這是一個生成式框架，將變化檢測重新定義為透過修正流在潛在空間中合成變化遮罩。ChangeFlow由結構化但輕量級的條件訊號引導，其隨機設計自然支援基於取樣的預測集成。具體而言，彙總多個預測的變化遮罩能提升穩健性，而樣本間的一致性則提供實用的信心估計，凸顯模糊區域。在四個基準測試中，ChangeFlow平均F1分數達80.4%，較先前最佳方法平均提升1.3個百分點，同時推論速度與近期強基線相當。專案頁面：https://blaz-r.github.io/changeflow_cd

English

Remote sensing change detection (RSCD) aims to localise changes between two images of the same geographic region. In practice, change masks often follow region-level annotation conventions rather than purely local appearance differences, making them context-dependent and occasionally ambiguous. Most state-of-the-art methods utilise per-pixel discriminative classification, which produces a single prediction per input and fails to explicitly model the changed region as a coherent whole. A natural alternative is generative formulation, which can model a distribution of plausible masks, enabling sampling to capture ambiguity and encourage global consistency. However, existing generative RSCD approaches typically lag behind strong discriminative baselines due to the high computational cost of pixel-space generation and the complexity of their conditioning mechanisms. To address the limitations of prior discriminative and generative methods, we propose ChangeFlow, a generative framework that reformulates change detection as the synthesis of a change mask in latent space via rectified flow. ChangeFlow is guided by a structured yet lightweight conditioning signal, and its stochastic design naturally supports sampling-based prediction ensembling. Namely, aggregating multiple predicted change masks improves robustness, while sample agreement provides a practical confidence estimation that highlights ambiguous regions. Across four benchmarks, ChangeFlow achieves an average F1 of 80.4\%, improving by 1.3 points on average over the previous best method, while maintaining inference speed comparable to recent strong baselines. Project page: https://blaz-r.github.io/changeflow_cd