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

Hugging Face Daily Papers Papers
change-detection remote-sensing generative-framework rectified-flow latent-space

Summary

ChangeFlow presents a generative framework for remote sensing change detection that synthesizes change masks in latent space using rectified flow, achieving improved accuracy and robustness through sampling-based prediction ensembling, with an average F1 of 80.4% across four benchmarks.

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
Original Article
View Cached Full Text

Cached at: 05/18/26, 02:26 PM

Paper page - ChangeFlow – Latent Rectified Flow for Change Detection in Remote Sensing

Source: https://huggingface.co/papers/2605.15375

Abstract

ChangeFlow presents a generative framework for remote sensing change detection that synthesizes change masks in latent space using rectified flow, achieving improved accuracy and robustness through sampling-based prediction ensembling.

Remote sensingchange 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 isgenerative 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 reformulateschange detectionas the synthesis of achange maskinlatent spaceviarectified flow. ChangeFlow is guided by a structured yet lightweightconditioning signal, and its stochastic design naturally supportssampling-based prediction ensembling. Namely, aggregating multiple predictedchange 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

View arXiv pageView PDFProject pageGitHub3Add to collection

Get this paper in your agent:

hf papers read 2605\.15375

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.15375 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.15375 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.15375 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Recursive Flow Matching

Hugging Face Daily Papers

Introduces Recursive Flow Matching (RecFM), a generative framework for forecasting complex spatiotemporal dynamics that achieves high fidelity with fewer steps and improved accuracy and speed, including up to 20x speedup over diffusion-based emulators.

Asymmetric Flow Models

Hugging Face Daily Papers

Asymmetric Flow Modeling (AsymFlow) restricts noise prediction to low-rank subspaces for efficient high-dimensional flow-based generation, achieving state-of-the-art results on ImageNet and text-to-image tasks by fine-tuning from latent flow models.

SDFlow: Similarity-Driven Flow Matching for Time Series Generation

arXiv cs.AI

This paper introduces SDFlow, a similarity-driven flow matching framework for time series generation that addresses exposure bias in autoregressive models. It achieves state-of-the-art performance and inference speedups by operating in the frozen VQ latent space with low-rank manifold decomposition.

SRT: Super-Resolution for Time Series via Disentangled Rectified Flow

arXiv cs.LG

This paper proposes SRT (Super-Resolution for Time Series), a framework that reconstructs high-resolution temporal patterns from low-resolution inputs using a disentangled rectified flow approach. The method decomposes input into trend and seasonal components, applies implicit neural representation for resolution alignment, and introduces cross-resolution attention to generate fine-grained details, achieving state-of-the-art performance on multiple datasets.