Unsupervised Domain Adaptation for Enhanced Radiometer Image Precipitation Estimation using Conditional Flow Matching

Victor Enescu; Assaad Zeghina; Matthieu Meignin; Nicolas Viltard; Cécile Mallet

cs.CVOct 1, 2026

Unsupervised Domain Adaptation for Enhanced Radiometer Image Precipitation Estimation using Conditional Flow Matching

Authors: Victor Enescu, Assaad Zeghina, Matthieu Meignin, Nicolas Viltard, Cécile Mallet

Organizations: LATMOS/IPSL, UVSQ Universit´e Paris-Saclay, Sorbonne Universit´e, CNRS Guyancourt, France · ARCHES Team, Inria Paris, France

Abstract

Deep generative networks have recently achieved unprecedented performance in precise image and video editing using sophisticated textual prompts. However, the effectiveness of such models heavily depends on access to very large supervised and annotated image datasets, which can be very difficult to obtain. This is particularly true for satellite instruments, which very rarely overlap with labelled data, and suffer from domain shifts in the rare occasions they do. In this paper, we investigate the potential of flow matching models for unsupervised domain adaptation of satellite radiometer images. Our main contribution is a novel unsupervised method that achieves precise domain alignment by leveraging parts of the deterministic ordinary differential equations in flow matching models, conditioned on different satellite instruments. A key strength of our approach is its ability to preserve essential information while adapting across any domains since the perturbations are in theory bijective. Extensive experiments conducted on the GPM-Core constellation show the benefit of our conditional domain adaptation, particularly in improving rain precipitation estimation from radiometer imagery.

Figures & tables

Figure 1 : Overview of the proposed conditional flow matching framework. During training (left), the model learns to map both GMI and F18 images to a standard Gaussian distribution using domain-specific labels. During inference (right), domain adaptation is performed in two steps: first, an F18 image is noised by running the forward ODE from $t_{0}$ to $t_{1}$ with the associated label conditioning ( ${\mathbf{y}}_{i}$ in Eq. 8 ). Afterwards, the reverse ODE is solved from $t_{1}$ to $t_{0}$ using the GMI conditioning parameters instead ( ${\mathbf{y}}_{j}$ in Eq. 8 ), so as to obtain the adapted image in the GMI domain (this corresponds to $\bar{{\mathbf{x}}}_{i\rightarrow j}$ in Eq. 8 ).

Instrument	Central Frequency	Polarisation	Bandwidth	Pixel Size (km)
GMI	36.6GHz	V, H	1000 MHz	$6.0\times 13.4$
SSMI/S	37.0GHz	V, H	1580 MHz	$25.0\times 12.5$
GMI	89.0GHz	V, H	6000 MHz	$3.0\times 13.4$
SSMI/S	91.6GHz	V, H	2829 MHz	$12.5\times 12.5$

Table 1 : Technical information on the considered GMI and SSMI/S, grouped by frequency bands and polarisations (V vertical, H horizontal).

RR in mm/h	Model	RMSE	MAE	ME
$[0.0,0.1)$	F18 Baseline	0.074	0.011	0.011
	SDEdit [ 13 ]	0.147	0.019	0.019
	DDIB [ 19 ]	0.102	0.012	0.012
	CycleGAN [ 16 ]	0.101	0.016	0.016
	Ours	0.071	0.007	0.007
$[0.1,2.5)$	F18 Baseline	0.638	0.412	-0.267

Table 2 : RMSE, MAE, and ME for different rain rate intervals.

Panel of Figure 2 — Figure 2 : Comparison of precipitation estimates (mm/h) obtained using DRAIN on the ground-truth GMI ( 2(a) ), the unadapted F18 baseline ( 2(b) ), our method ( 2(c) ), and the comparative methods DDIB [ 19 ] , SDEdit [ 13 ] , and CycleGAN [ 16 ] ( 2(d) , 2(e) , 2(f) ). Results are shown for the 2019-02-06 overpass all projected on GMI pixels. Note that perfect collocation is unattainable due to differences in observation angles, acquisition times, and pixel counts, which may cause discrepancies despite “perfect” adaptation.

Panel of Figure 3 — Figure 3 : Comparison between the original F18 images (37GHz-V channel) ( 3(a) ), with the adapted images generated using our proposed method ( 3(b) ), as well as the comparative methods DDIB [ 19 ] ( 3(c) ), SDEdit [ 13 ] ( 3(d) ), and CycleGAN [ 16 ] ( 3(e) ).

Explore similar work

Flow Matching for Convective-Scale Precipitation Downscaling
May 29, 2026Tom WetherellDynamical Downscaling Generative Models
Met Office
FlowForm: Synergizing Fluid Physics with Topological Consistency for Satellite Flood Synthesis
Aug 4, 2026Zhang Weihui, Wang Ruizhi, Xu Hongye +3Flood Detection Satellite Imagery
Zhejiang University, Zhejiang, China
ICM: Intra-class Mixing for Domain Adaptation in Adverse Weather
Sep 23, 2026Boying Li, Chang Liu, Britta Ayano Wilde +4Domain Adaptation Efficient Semantic Segmentation

May 29, 2026, physics.ao-phJ/K move · Enter open · S save

Flow Matching for Convective-Scale Precipitation Downscaling

Tom Wetherell

Met Office

Dynamical Downscaling Generative Models

Open paper AskPDF arXiv

Generative machine learning is an increasingly important complement to dynamical downscaling for producing high-resolution precipitation projections, with diffusion models currently the leading approach. Flow matching is a related generative framework that has recently achieved strong results across image, video and other domains, and shown early promise for downscaling. We train a flow matching model to map daily precipitation from 8 km to 2 km over a convective-scale domain centred on Singapore, and benchmark it against CPMGEM, a score-based diffusion model. Flow matching achieves consistently better spatial skill: higher fractions skill score at every precipitation threshold and neighbourhood scale tested, and tighter structure and amplitude components of the SAL score with comparable location skill. However, flow matching underestimates the upper tail of the precipitation distribution, resulting in a dry bias in the climatological mean. These results suggest that flow matching is a competitive generative framework for convective-scale precipitation downscaling, particularly well suited to capturing spatial structure.

Abstract

Explore similar work

Flow Matching for Convective-Scale Precipitation Downscaling

FlowForm: Synergizing Fluid Physics with Topological Consistency for Satellite Flood Synthesis

ICM: Intra-class Mixing for Domain Adaptation in Adverse Weather