ChangeFlow -- Latent Rectified Flow for Change Detection in Remote Sensing

TL;DR

ChangeFlow introduces a novel generative approach using latent rectified flow for more coherent and robust change detection in remote sensing images, outperforming previous methods in accuracy and efficiency.

Contribution

It proposes ChangeFlow, a latent space generative framework that models change masks as a distribution, enabling sampling, global consistency, and improved robustness in remote sensing change detection.

Findings

Achieves an average F1 score of 80.4%, surpassing previous best by 1.3 points.

Maintains inference speed comparable to strong discriminative baselines.

Supports sampling-based prediction ensembling for better robustness and confidence estimation.

Abstract

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